Every customer base is unique. Your ability to win your own marketing victories depends on understanding the users visiting your website each and every day. If you’ve spent any amount of time on The Daily Egg or similarly focused blogs, you know that thorough testing is paramount to your web business’ success. You might be intuitive (or lucky) enough to correctly read your customers on any given campaign, but over the long run, success WILL depend on data analysis. Despite the hundreds of voices claiming that marketing is an art, you’ve heard of Neil Patel, Avinash Kaushik, Danny Sullivan, and…
Year round Liz and Bill Farrell, a husband and wife team, work the dirt at Fat Stone Farm in Lyme, Connecticut with their two kids.
After making the move from cubicles to the great outdoors, the Farrells realized they loved growing fresh food, and creating healthy, farm-grown products ranging from elderberry apple shots to their own maple syrup.
Now—when you think of a typical farmer’s marketing strategy—you might picture a hand-painted sign at a local market, but Liz and Bill run a digital elderberry empire.
The couple started as Shopify merchants and then partnered with digital agency Webistry to take their business to new heights. They wanted to see better return on ad spend, and prepare for winter (their best sales season of the year).
It was a perfect partnership from the start, but nobody could predict that a combo of Unbounce landing pages, popups, a Shopify integration, and near obsessive audience building and retargeting on Facebook would:
Lower cost-per-acquisition from $145 (at its highest) to just $1.55(!) for the company’s Elderberry Apple Shots and DIY Gummy kits.
Increase return on ad spend for their Elderberry Apple Shot campaigns from 1.66X to an incredible 33.12X.
Deliver a cost-per-lead for a sweepstakes campaign of just $0.51.
See sweepstakes conversion rates from ad click to entry of up to 79.55%.
And garner Facebook relevance scores of 8s and 9s.
Overall, with Webistry’s help, Fat Stone Farm tripled ROAS in just five months (December ‘16–April ‘17), and—via continued optimizations—reached returns of 33X over a year and three months.
Here’s their epic ecommerce story, and the paid media tactics that could work for you too.
Winter is Coming
Historically, sales of Fat Stone Farm’s Elderberry Apple Shots go up in winter to help fight off flu season. So in early winter 2016, Bill and Liz approached Jonathan Naccache, Co-Founder at Webistry to prep some advertising.
The agency discovered that they couldn’t look to AdWords for a huge win. The search volume for elderberries or related products wasn’t super high, and this approach simply wasn’t scalable. Instead, they needed to generate extremely targeted custom audiences on Facebook (which can be difficult because prospects on Facebook aren’t necessarily screaming about their love of elderberry – these prospects need to be uncovered).
In Webistry’s approach, each ad campaign would target a group of interests that could coincide with elderberry products. They’d target Facebook users who’s interests included: alternative medicine, natural remedies, homesteading, or those engaging with popular health blogs like Mother Earth News and Wellness Mama.
It took a lot of research, and as Jonathan says this is where the agency advantage comes into play: “having access to several strategic minds, resources and thorough research gets you a significant edge right off the start.”
The First Ad to Shopify Landing Page Combo
From December 2016 to January 2017, Webistry ran campaigns on Facebook targeting each of the audience segments they’d identified might be interested in the elderberry shots.
Here’s an example of some of the ads (corresponding to fall and winter seasons):
Pictured above: the ads Webistry ran to the associated landing page.
All elderberry apple shots ads led to this Unbounce-built PPC landing page, which converts at 4.7% (lifetime average conversion rate). A conversion in this case was a purchase via the ‘add to cart’ button):
Notice the benefit copy from the ad headlines is carried through to the Unbounce landing page. Click to see a larger version.
Two months into this campaign, return on ad spend was 1.66X, and cost-per-purchase was fluctuating between $19 and $145. Jonathan knew they could improve upon these early results and began targeting audiences of vegetarians, vegans, healthy eating affectionados, and homesteaders.
And so, in April 2017 the agency launched a new landing page campaign for smoothie lovers.
The idea was to position the elderberry product as the ideal ingredient to add to a smoothie. Here’s a sample ad used to launch this campaign:
And of course, the landing page this ad pointed to:
This beautiful landing page converts traffic to purchase at 9.44%
Beyond driving sales, the agency realized there was potential for lead capture here too (as a means of remarketing to especially interested prospects later), so they added an on-exit popup to this page. It offered up a free smoothie recipe book and integrated with a Mailchimp autoresponder.
With an 18% conversion rate, here’s the popup built in Unbounce:
The smoothie campaign helped drop cost-per-purchase down to ~$9.65, and Bill and Liz saw a return on ad spend of 3X from their initial investment after just five months of executing this strategy.
This was great, but Webistry wanted to help Fat Stone Farm stay present in their potential buyer’s world year round. They could lie low waiting for winter again all summer, or they could start developing highly refined retargeting and lookalike audiences to reach all year long.
The Sweepstakes That Raised the Stakes
In May 2017, continuing in the off-season, it was time to start preparing for their next winter. Fat Stone Farm was seeing major benefits from refined Facebook audience targeting, so Jonathan and the team extended this strategy with sweepstakes.
They used weekly sweepstakes as a means to gauge and track prospect’s interest in the products, then later in the winter, they created Facebook lookalike and retargeting audiences to get in front of similar groups of interested people regularly.
As Jonathan shares, this allowed the team to generate even better target audiences:
“Our goal was to create campaigns that helped us measure different levels of interest, and to identify these audiences by tracking every event with a pixel. We had a drip campaign setup, and non-winners of the first sweepstakes were given access to a second sweepstakes.”
That is – those who didn’t win each week were offered access to another sweepstakes prize (either the breakfast pack or gummy pack product). This helped introduce prospects to other Fat Stone Farm products and gauge interest for these versus a complementary offering like smoothies.
Here’s a sample ad for the sweepstakes:
And here’s the first landing page touchpoint:
Click the image above for a closer look.
If you didn’t win, you might be sent a second offer in the sweepstakes, with a chance to win an Elderberry Gummy Kit via the landing page below:
Click the image above to see the full landing page.
Of the people who clicked through on the Facebook ad and reached the first landing page above, 18.79% converted. Moreover, of the people who did not win the first sweepstakes, but clicked through the email announcing the second sweepstakes, 79.55% converted via the landing page.
Hot tip: Webistry embedded a third party tool called ViralSweep on these pages. It’s a sweepstakes application to help manage entries, select a winner at random, and allows people to win bonus entries by referring friends via social.
Not only did this campaign collect over 15,000 relevant leads that Fat Stone Farm could remarket to year long with terrific offers, but it reduced cost-per-lead down to a mere $0.51.
Which brings us to…
Winter Season, 2018
After all the ad testing, landing page alterations, and lessons along the way, Webistry re-launched the sales campaigns using six months of audience-segmented data.
They launched the gummy kits as a standalone product landing page (vs. the sweepstakes page) and continued to sell the Elderberry Apple Shots. The best part? From January to March 2018 Webistry achieved the highest return on ad spend for Bill and Liz since starting to work with them: a whopping 33.12X.
Additionally, this season they saw the lowest cost-per-acquisition of just $1.55.
As Bill Ferrell says of the partnership with Webistry:
“These guys are worth every penny. Excellent results (very high CTRs, good CPA, [and] lots of new customers!). The Webistry co-founders are hands-on, creative, and keep tweaking throughout. Their attentiveness to the campaigns and my crazy ideas have exceeded my expectations month after month.”
WordPress Local Development For Beginners: From Setup To Deployment
When first starting out with WordPress, it’s very common to make any changes directly on your live site. After all, where else would you do it? You only have that one site, so when something needs changing, you do it there.
However, this practice has several drawbacks. Most of all that it’s very public. So, when something goes seriously wrong, it’s quite noticeable for people on your site.
Preventing Common Mistakes
When creating free or premium WordPress themes, you’re bound to make mistakes. Find out how you can avoid them in order to save yourself time and focus on actually creating themes people will enjoy using. Read more →
It’s ok, don’t feel bad. Most WordPress beginners have done this at one point or another. However, in this article, we want to show you a better way: local WordPress development.
What that means is setting up a copy of your website on your local hard drive. Doing so is incredibly useful. So, below we will talk about the benefits of building a local WordPress development environment, how to set one up and how to move your local site to the web when it’s ready.
This is important, so stay tuned!
The Benefits Of Local WordPress Development
Before diving into the how, let’s have a look at the why. Using a local development version of WordPress offers many benefits.
We already mentioned that you no longer have to make changes to your live site with all the risks involved with that. However, there is more:
Test themes and plugins With a local copy of your site, you can try out as many themes and plugin combinations as you want without risking taking your live site out due to incompatibilities.
Update safely Another time when things are prone to go wrong are updates. With a local environment, you can update WordPress core and components to see if there are any problems before applying the updates to your live site.
Independent of an online connection With your WordPress site on your computer, you can work on it without being connected to the Internet. Thus, you can get work done even if there is no wifi.
High performance/low cost Because site performance is not limited by an online connection, local sites usually run much faster. This makes for a better workflow. Also, as you will see, you can set it all up with free software, eliminating the need for a paid staging area.
Sounds good? Then let’s see how to make it happen.
How To Set Up A Local Development Environment For WordPress
In this next part, we will show you how to set up your own local WordPress environment. First, we will go over what you need to do and then how to get it right.
Tools You’ll Need
In order to run, WordPress needs a server. That’s true for an online site as well as a local installation. So, we need to find a way to set one up on our computer.
That server also needs some software WordPress requires to work. Namely, that’s PHP (the platform’s main programming language) and MySQL for the database. Plus, it’s nice to have a MySQL user interface like phpMyAdmin to make handling the database more convenient.
Finally, it’s useful to have some developer tools to analyze and debug your site, for example, to look at HTML and CSS. The easiest way is to use Chrome or Firefox (read our article on Firefox’s DevTools), which have extensive functionality like that built in.
We have several options at our disposal to set up local server environments. Some of the most well known are DesktopServer, Vagrant, and Local by Flywheel. All of these contain the necessary components to set up a local server that WordPress can run on.
For this tutorial we will use XAMPP. The name is an acronym and stands for “cross platform, Apache, MySQL, PHP, Perl”. If you have been paying attention, you will notice that we earlier noted MySQL and PHP as essential to running a WordPress website. In addition, Apache is an open source solution for creating servers. So, the software contains everything we need in one neat package. Plus, as “cross platform” suggests, XAMPP is available for both Windows, Mac and Linux computers.
Installing XAMPP pretty much works like every other piece of software.
Run the installer (note that you might get a warning about running unknown software, allow to continue).
When asked which components to install, make sure that Apache, MySQL, PHP, and phpMyAdmin are active. The rest is usually unnecessary, deactivate it unless you have good reason not to.
Choose the location to install. Make sure it’s easy to reach as that’s where your sites will be saved and you will probably access them often.
You can disregard the information about Bitnami.
Choose to start the control panel right away at the end.
Open the .dmg file
Double click on the XAMPP icon or drag it to applications folder
That’s it, well done!
After the installation is complete, the control panel starts. Should your operating system ask for Firewall permissions, make sure to allow XAMPP for private networks, otherwise, it won’t work.
From the panel, you can start Apache and MySQL by clicking the Start buttons on the respective rows. If you run into problems with programs that use the same ports as XAMPP, quit those programs and try to restart the XAMPP processes. If the problem is with Skype, there is a permanent solution by disabling the ports under Tools → Options → Advanced → Connections.
Under Config, you can also enable automatic start for the components you need.
After that, it’s time to test your local server. For that, open your browser, and go to http://localhost.
If you see the following screen, everything works as it should. Well done!
Installing WordPress Locally
Now that you have a local server, you can install WordPress in the same way that you do on a web server. The only difference: everything is done on your hard drive, not an FTP server or inside a hosting provider’s admin panel.
Once that is done and you want to install WordPress, you can do so via the htdocs folder inside your installation of XAMPP. There, simply create a new directory, download the latest version of WordPress, unpack the files and copy them into the new folder. After that, you can start the installation by going to http://localhost/newdirectoryname.
When you are satisfied, you can then move the website from a local installation to live environment. That’s what we will talk about next.
How To Deploy Your Site With A Plugin
Alright, once your local site is up to your liking, it’s time to get it online. Just a quick note: if you want to get a copy of your existing live site to your hard drive, you can use the same process as described below only in reverse. The overall principles stay the same.
Either way, we can do this manually or via plugin and there are several solutions available. For this tutorial, we will use Duplicator. I found it to be one of the most convenient free solutions and, as you will see, it makes things really easy.
1. Install Duplicator
Like every other WordPress plugin, you first need to install Duplicator in order to use it. For that, simply go to Plugins → Add New. In the search box, input the plugin name and hit enter, it should be the first search result.
Click Install Now and activate once it’s done.
2. Export Site
When Duplicator is active on your site, it adds a new menu item in the WordPress dashboard. Clicking it gets you to this screen.
Here, you can create a so-called package. In Duplicator that means a zipped up version of your site, database, and an installation file. To get started, simply click on Create New.
In the next step, you can enter a name for your package. However, it’s not really necessary unless you have a specific reason.
Under Storage, Archive, and Installer are options to determine where to save the archive (works for the Pro version only), exclude files or database tables from the migration and input the updated database credentials and new URL.
In most cases you can just leave everything as is, especially because Duplicator will attempt to fill in the new credentials automatically later. So, for now just hit Next.
After that, Duplicator will run a test to see if there are any problems.
Unless something major pops up, just click on Build to begin building the package. This will start the backup process.
At the end of it, you get the option to download the zip file and installer with the click on the respective buttons or with the help of the One-Click Download.
Do so and you are done with this part of the process.
3. Upload And Deploy Files On Your Server
If all of this has gone down without a hitch, it’s now time to upload the generated files to their new home. For that, log into your FTP server and browse to your new site’s home directory. Then, start uploading the files we generated in the last step.
Once done, you can start the installation process by inputting http://yoursite.com/installer.php into the browser bar.
In the first step, the plugin checks the archive’s integrity and whether it can deploy the site in the current environment. You also get some advanced options that you are welcome to ignore at the moment. However, make sure to check the box where it says “I have read and accept all terms and notices,” before clicking Next.
Your site is now being unpacked. After that, you get to the screen where it’s time to input the database information.
The plugin can either create a new database (if your host allows it) or use an existing one. For the latter option, you need to set up the database manually beforehand. Either way, you need to input the database name, username, and password to continue. Duplicator will also use this information to update the wp-config.php file of your site so that it can talk to the new database. Click Test Database to see if the connection works. Hit Next to start the installation of the database.
Once Duplicator is done with that, the final step is to confirm the details of your old and new site.
That way, Duplicator is able to replace all mentions of your old URL in the database with the new one. If you don’t, your site won’t work properly. If everything is fine, click the button that says Next.
4. Finishing Up
Now, there are just a few more things to do before you are finished. The first one is to check the last page of the setup for any problems encountered in the deployment.
The second is to log into your new site (you can do so via the button). Doing so will land you in the Duplicator menu.
Here, be sure to click on Remove Installation Files Now! at the top. This is important for security reasons.
And that’s it, your site should now be successfully migrated. Well done! You have just mastered the basics of local WordPress development.
Quick Note: Updating Your Database Information Manually
Should the Duplicator plugin for some reason be unable to update wp-config.php with the new database information, your site won’t work and you will see a warning that says “unable to establish database connection”.
In that case, you need to change the information manually. Do this by finding wp-config.php in your WordPress installation’s main folder. You can access it via FTP or a hosting backend like cPanel. Ask your provider for help if you find yourself unable to locate it on your own.
Edit the file (this might mean downloading, editing and re-uploading it) and find the following lines:
/** The name of the database for WordPress */
/** MySQL database username */
/** MySQL database password */
/** MySQL hostname */
Update the information here with that of your new host (by replacing the info between the ‘ ‘), save the file and move it back to your site’s main directory. Now everything should be fine.
WordPress Local Development In A Nutshell
Learning how to install WordPress locally is super useful. It enables you to make site changes, run updates, test themes and plugins and more in a risk-free environment. In addition to that, it’s free thanks to open source software.
Above, you have learned how to build a local WordPress environment with XAMPP. We have led you through the installation process and explained how to use the local server with WordPress. We have also covered how to get your local site online once it’s ready to see the light of day.
Hopefully, your takeaway is that all of this is pretty easy. It might feel overwhelming as a beginner at first, however, using WordPress locally will become second nature quickly. Plus, the benefits clearly outweigh the learning process and will help you take your WordPress skills to the next level.
What are your thoughts on local WordPress development? Any comments, software or tips to share? Please do so in the comments section below!
Measuring Websites With Mobile-First Optimization Tools
Performance on mobile can be particularly challenging: underpowered devices, slow networks, and poor connections are some of those challenges. With more and more users migrating to mobile, the rewards for mobile optimization are great. Most workflows have already adopted mobile-first design and development strategies, and it’s time to apply a similar mindset to performance.
In this article, we’ll take a look at studies linking page speed to real-world metrics, and discuss the specific ways mobile performance impacts your site. Then we’ll explore benchmarking tools you can use to measure your website’s mobile performance. Finally, we’ll work with tools to help identify and remove the code debt that bloats and weighs down your site.
How would you design a responsive car configurator? How would you deal with accessibility, navigation, real-time previews, interaction and performance? Let’s figure it out. Read article →
Why Performance Matters
The benefits of performance optimization are well-documented. In short, performance matters because users prefer faster websites. But it’s more than a qualitative assumption about user experience. There are a variety of studies that directly link reduced load times to increased conversion and revenue, such as the now decade-old Amazon study that showed each 100ms of latency led to a 1% drop in sales.
Page Speed, Bounce Rate & Conversion
In the data world, poor performance leads to an increased bounce rate. And in the mobile world that bounce rate may occur sooner than you think. A recent study shows that 53% of mobile users abandon a site that takes more than 3 seconds to load.
That means if your site loads in 3.5 seconds, over half of your potential users are leaving (and most likely visiting a competitor). That may be tough to swallow, but it is as much a problem as it is an opportunity. If you can get your site to load more quickly, you are potentially doubling your conversion. And if your conversion is even indirectly linked to profits, you’re doubling your revenue.
SEO And Social Media
Beyond reduced conversion, slow load times create secondary effects that diminish your inbound traffic. Search engines already use page speed in their ranking algorithms, bubbling faster sites to the top. Additionally, Google is specifically factoring mobile speed for mobile searches as of July 2018.
Social media outlets have begun factoring page speed in their algorithms as well. In August 2017, Facebook announced that it would roll out specific changes to the newsfeed algorithm for mobile devices. These changes include page speed as a factor, which means that slow websites will see a decline in Facebook impressions, and in turn a decline in visitors from that source.
Search engines and social media companies aren’t punishing slow websites on a whim, they’ve made a calculated decision to improve the experience for their users. If two websites have effectively the same content, wouldn’t you rather visit one that loads faster?
Many websites depend on search engines and social media for a large portion of their traffic. The slowest of these will have an exacerbated problem, with a reduced number of visitors coming to their site, and over half of those visitors subsequently abandoning.
If the prognosis sounds alarming, that’s because it is! But the good news is that there are a few concrete steps you can take to improve your page speeds. Even the slowest sites can get “sub three seconds” with a good strategy and some work.
Profiling And Benchmarking Tools
Before you begin optimizing, it’s a good idea to take a snapshot of your website’s performance. With profiling, you can determine how much progress you will need to make. Later, you can compare against this benchmark to quantify the speed improvements you make.
There are a number of tools that assess a website’s performance. But before you get started, it’s important to understand that no tool provides a perfect measurement of client-side performance. Devices, connection speeds, and web browsers all impact performance, and it is impossible to analyze all combinations. Additionally, any tool that runs on your personal device can only approximate the experience on a different device or connection.
In one sense, whichever tool you use can provide meaningful insights. As long as you use the same tool before and after, the comparison of each should provide a decent snapshot of performance changes. But certain tools are better than others.
In this section, we’ll walk through two tools that provide a profile of how well your website performs in a mobile environment.
Note: If can be difficult to benchmark an entire site, so I recommend that you choose one or two of your most important pages for benchmarking.
One of the more useful tools for profiling mobile performance is Google’s Lighthouse. It’s a nice starting point for optimization since it not only analyzes page performance but also provides insights into specific performance issues. Additionally, Lighthouse provides high-level suggestions for speed improvements.
Lighthouse is available in the Audits tab of the Chrome Developer Tools. To get started, open the page you want to optimize in Chrome Dev Tools and perform an audit. I typically perform all the audits, but for our purposes, you only need to check the ‘Performance’ checkbox:
Lighthouse focuses on mobile, so when you run the audit, Lighthouse will pop your page into the inspector’s responsive viewer and throttle the connection to simulate a mobile experience.
When the audit finishes, you’ll see an overall performance score, a timeline view of how the page rendered over time, as well as a variety of metrics:
It’s a lot of information, but one report to emphasize is the first meaningful paint, since that directly influences user bounce rates. You may notice that the tool doesn’t even list the total load time, and that’s because it rarely matters for user experience.
Mobile users expect a first view of the page very quickly, and it may be some time before they scroll to the lower content. In the timeline above, the first paint occurs quickly at 1.3s, then a full above-the-fold content paint occurs at 3.9s. The user can now engage with the above-the-fold content, and anything below-the-fold can take a few seconds longer to load.
Lighthouse’s first meaningful paint is a great metric for benchmarking, but also take a look at the opportunities section. This list helps to identify the key problem areas of your site. Keep these recommendations on your radar, since they may provide your biggest improvements.
While Lighthouse provides great insights, it is important to bear in mind that it only simulates a mobile experience. The device is simulated in Chrome, and a mobile connection is simulated with throttling. Actual experiences will vary.
Additionally, you may notice that if you run the audit multiple times, you will get different reports. That’s again because it is simulating the experience, and variances in your device, connection, and the server will impact the results. That said, you can still use Lighthouse for benchmarking, but it is important that you run it several times. It is more relevant as a range of values than a single report.
In order to get an idea of how quickly your page loads in an actual mobile device, use WebPageTest. One of the nice things about WebPageTest is that it tests on a variety of real devices. Additionally, it will perform the test a number of times and take the average to provide a more accurate benchmark.
To get started, navigate to WebPageTest.org, enter the URL for the page you want to test and then select the mobile device you’d like to use for testing. Also, open up the advanced settings and change the connection speed. I like testing at Fast 3G, because even when users are connected to LTE the connection speed is rarely LTE (#sad):
After submitting the test (and waiting for any queue), you’ll get a report on the speed of the page:
The summary view consists of a short list of metrics and links to timelines. Again, the value of the start render is more important than the load time. The first byte is useful for analyzing the server response speed. You can also dig into the more in-depth reports for additional insights.
Now that you’ve profiled your page in Lighthouse and WebPageTest, it’s time to record the values. These benchmarks will provide a useful comparison as you optimize your page. If the metrics improve, your changes are worthwhile. If they stay static (or worse decline), you’ll need to rethink your strategy.
Lighthouse results are simulated which makes it less useful for benchmarking and more useful for in-depth reports and optimization suggestions. However, Lighthouse’s performance score and first meaningful paint are nice benchmarks so run it a few times and take the median for each.
WebPageTest’s values are more reliable for benchmarking since it tests on real devices, so these will be your primary benchmarks. Record the value for the first byte, start to render, and overall load time.
Now that you’ve assessed the performance of your site, let’s take a look at a tool that can help reduce the size of your files. This tool can identify extra, unnecessary pieces of code that bloat your files and cause resources to be larger than they should.
Additionally, certain resources are intended to be cached and then used throughout multiple pages, such as a site-wide stylesheet. Site-wide resources often make sense, but how can you tell when a stylesheet is mostly underused?
The Coverage Tab
Fortunately, Chrome Developer Tools has a tool that helps assess the bloat in files: The Coverage tab. The Coverage tab analyzes code coverage as you navigate your site. It provides an interface that shows how much code in a given CSS or JS file is actually being used.
To access the Coverage tab, open up Chrome Developer Tools, and click on the three dots in the top right. Navigate to More Tools > Coverage.
Next, start instrumenting coverage by clicking the reload button on the right. That will reload the page and begin the code coverage analysis. It brings up a report similar to this:
Here, pay attention to the unused bytes:
This UI shows the amount of code that is currently unused, colored red. In this particular page, the first file shown is 73% bloat. You may see significant bloat at first, but it only represents the current render. Change your screen size and you should see the CSS coverage go up as media queries get applied. Open any interactive elements like modals and toggles, and it should go up further.
Once you’ve activated every view, you will have an idea of how much code you are actually using. Next, you can dig into the report further to find out just which pieces of code are unused, simply click on one of the resources and look in the main window:
In this CSS file, look at the highlights to the left of each ruleset; green indicates used code and red indicates bloat. If you are building a single page app or using specialized resources for this particular page, you may be inclined to go in and remove this garbage. But don’t be too hasty. You should definitely remove dead code, but be careful to make sure that you haven’t missed a breakpoint or interactive element.
In this article, we’ve shown the quantitative benefits of optimizing page speed. I hope you’re convinced, and that you have the tools you need to convince others. We’ve also set a minimum goal for mobile page speed: sub three seconds.
To hit this goal, it’s important that you prioritize the highest impact optimizations first. There are a lot of resources online that can help define this roadmap, such as this checklist. Lighthouse can also be a great tool for identifying specific issues in your codebase, so I encourage you to tackle those bottlenecks first. Sometimes the smallest optimizations can have the biggest impact.
You might not wake up each morning thinking about data privacy and security but, like it or not, Facebook’s recent move makes it an issue you can’t dismiss. Long before Mark Zuckerberg sat before congress in the face of the Cambridge Analytica scandal, explaining how Facebook uses personal data, the European Union started getting especially serious about data protection and privacy.
And so, on May 25 2018, the EU’s General Data Protection Regulation (GDPR) goes into effect.
In a nutshell, the GDPR legislation gives everyone in the EU greater privacy rights, and introduces new rules for marketers and software providers to follow when it comes to collecting, tracking, or handling EU-based prospects’ and customers’ personal data.
Moreover, the GDPR applies to anyone who processes or stores data of those in the EU (i.e. you don’t need to be physically located in Europe for this to apply to your business and can incur fines up to 4% of your annual global turnover or €20 million [whichever is greater] for non-compliance).
But Beyond Potential Fines, Here’s Why You Need to Care
On Tuesday April 3rd, Zuckerberg said that Facebook had no plans to extend the GDPR regulations globally to all Facebook users. But, fast-forward a few weeks later and Facebook completely changed its tune, now planning to extend Europe’s GDPR standards worldwide.
This move sets a precedent, showing all of us that no matter where we are in the world, personal data and privacy laws aren’t optional. Compliance is table stakes.
If you’re located in Europe, process lead and customer data from Europe — or just happen to believe in high standards for data privacy and security, this post will help you navigate:
What Unbounce has done to become GDPR compliant, and
Some of what you need to do to make sure your landing pages, sticky bars, and popups adhere to the new rules.
Note: This post isn’t the be-all-and-end-all on EU data privacy, nor is it legal advice. It’s meant to provide background information and help you better understand how you can use Unbounce in a GDPR compliant way.
Data Protection by Default for You and Your Customers
For several months now, Unbounce has been investing heavily in the necessary changes to be GDPR compliant as a conversion platform. We believe that to build trust and confidence with your customers, you need to make their privacy your priority.
As of the day of GDPR enforcement, you can be sure we’ve got your back when it comes to processing and storing your data within Unbounce, and giving you the tools you need to run compliant campaigns.
To see exactly what Unbounce has been doing, why it matters and where we’re at in development, check out our GDPR FAQ page.
But while we’re a GDPR compliant platform with privacy and security safeguards built into our business practices and throughout our platform, this is only part of the equation. There are still a few things you are responsible for to use Unbounce in a compliant way, including:
Obtaining consent from your visitors (lawful basis of processing)
Deleting personal data if requested (right to erasure)
Encrypting lead data at transit and in rest (using SSL) and
Signing a data processing addendum (DPA) with Unbounce
Here’s what you’re gonna want to watch for as you build landing pages, popups, and sticky bars.
Obtaining Consent From Your Visitors
Before collecting someone’s data the GDPR states you must first have a legal basis to do so. There are six lawful bases of processing under the GDPR, but if you’re a digital marketer, your use case will most likely fall into one of the following three:
Consent (i.e. opt-in)
Performance of a contract (eg. sending an invoice to a customer)
“Legitimate interest” (eg. Someone is an existing customer and you want to send them information related to a product or service they already have.)
If you are using Unbounce for lead gen, then you must gather consent via opt-in to collect, use, or store someone’s data. When building your landing pages in Unbounce, you can easily add an opt-in field to your forms with the Unbounce form builder:
Keep in mind: Your visitors must actively check your opt-in box to give consent. Pre-checked checkboxes are not a valid form of consent.
Related But Different: Cookies And The ePrivacy Regulation
In many posts you’ll see Europe’s ePrivacy regulations tied in with GDPR, but they are, in fact, two separate things. While the GDPR regulates the general use and management of personal data, cookie use is core to the ePrivacy regulation (which is why you’ll sometimes see it called the “cookie law”). ePrivacy regulations are still in the works, but it’s certain they will be about visitor consent to cookies on your site.
Last year Unbounce launched sticky bars (a discreet, mobile-friendly way to get more conversions), but they do double duty as a cookie bar, notifying your visitors about cookies.
You can design and publish a cookie bar using Unbounce’s built-in template, as shown below, embed the code across all of your landing pages using script manager, then promptly publish to every landing page you build in Unbounce. You can even have it appear all across your website.
Informing Visitors of Your Data Protection Policies
It’s not enough to just obtain consent, the GDPR also requires you to inform your customers and prospects what they are consenting to. This means that you need to provide easy access to your privacy and data protection policies (something Google AdWords has required for ages).
Sharing your privacy and data protection policies easily and transparently can help you earn the trust and confidence of your web visitors. Every visitor may not read through it with a fine tooth comb, but in a web littered with sketchy marketing practices, sharing your policies shows that you’re legit and that you have nothing to hide.
This is a great example of how doing right by your customers can also help you achieve your business goals.
The Right To Be Forgotten
At any point in time a customer or lead whose data you have collected can request that you erase any of their personal data you have stored. There are several grounds under which someone can make this request and the GDPR requires that you do so without “undue delay”.
As an Unbounce customer, simply submit an email request to our support team who will ensure that all information for a specific lead or a group of leads are deleted from our database.
As part of our ongoing commitment to supporting data privacy and security, we are inspecting alternate solutions to deletion requests, but you can rest assured that even as of today, we will fulfill deletion requests within the time limit enforced by the GDPR.
Preventing Unauthorized Access to Data
Unbounce has supported SSL encryption on landing pages for years, and we’re proud that we made this a priority for our customers before Google started calling out non-https pages as not secure and giving preferential treatment to secure pages.
Presently Unbounce customers can already adhere to the GDPR requirement to process all data securely.
When you build and publish your landing pages with Unbounce, you can force your web visitors to the secure (https) version of your pages, even if they accidentally navigate to the unsecure (http) version.
In the upper right corner you can toggle to force visitors to the secure HTTPS version of your page.
This forced redirect will ensure proper encryption of your visitor lead data in transit and at rest. And as an added bonus, it’ll keep you in Google’s good books and prevent ‘not secure’ warnings in Google Chrome.
Signing a Data Protection Addendum (DPA) With Unbounce
According to the GDPR, when you collect lead information with Unbounce, you are the data controller while Unbounce serves as your data processor. To comply with GDPR regulation when using a tool like a landing page builder or conversion platform, you need a signed DPA between you (the data controller) and the service provider (your data processor).
Without getting too deep into the weeds on this one, let me just say that if you’re using Unbounce, we’ve got you covered and that you can complete a form on our GDPR overview page to get your DPA by email.
Privacy = Trust = Great Marketing
At Unbounce we view data privacy and security as two cornerstones of great marketing. At their core they are about a positive user experience and can help make the internet a better place.
The GDPR puts more control in the hands of users to determine how their information is used. No one wants their personal data falling into the wrong hands or being used in malicious or intrusive ways. Confidence and trust in your brand is at stake when it comes to privacy, so we aren’t taking any chances. Using Unbounce as your conversion platform, you can assure your customers that you take their privacy and data security seriously.
Increased regulation around data privacy may provide short term challenges for marketers as we establish new norms, but long term they can provide a more positive experience for users — something we should always strive for as marketers.
It’s 2018, and CRO isn’t just a buzzword anymore! Over the past decade, savvy businesses have been growing by not only investing in traffic acquisition strategies, but also ensuring that visitors to their website are converting into customers.
At VWO, we understand how daunting and time-consuming CRO can seem, so we joined hands with HubSpot to bring you a DIY guide, which will help you learn and implement process-oriented CRO for your business.
In our experience of working with 5,000+ customers across the globe, we’ve seen that the journey from start to first few home runs in optimizing conversions usually takes 8 weeks.
Therefore, we’ve designed this guide to take you on a week-by-week journey on how you can lift your conversion rates in a methodical, sustainable manner. Here’s what the 8-week of conversion optimization journey will cover:
Understanding the goals and principles of CRO
Conducting a conversion rate audit for your website
Identifying areas of improvement in your conversion funnel
Conducting qualitative research into your visitor behavior
Constructing educated hypotheses and prioritizing these for testing
Choosing the right experiment and setting up your testing platform
Analyzing and learning from your A/B test results
Ensuring continuous growth through CRO
After you’ve followed this guide, you’ll be equipped with the know-hows to increase conversion rates time and again, instead of doing it just once.
What’s more, even if your company is young or on a shoestring budget, you would be able to effectively practice conversion optimization in-house, all by yourself.
Grab your copy of The Complete DIY Guide To Improving Conversions in 60 Dayshere.
Working Together: How Designers And Developers Can Communicate To Create Better Projects
Among the most popular suggestions on Smashing Magazine’s Content User Suggestions board is the need of learning more about the interaction and communication between designers and developers. There are probably several articles worth of very specific things that could be covered here, but I thought I would kick things off with a general post rounding up some experiences on the subject.
Given the wide range of skills held by the line-up at our upcoming SmashingConf Toronto — a fully live, no-slides-allowed event, I decided to solicit some feedback. I’ve wrapped those up with my own experience of 20 years working alongside designers and other developers. I hope you will add your own experiences in the comments.
Some tips work best when you can be in the same room as your team, and others are helpful for the remote worker or freelancer. What shines through all of the advice, however, is the need to respect each other, and the fact that everyone is working to try and create the best outcome for the project.
Working Remotely And Staying Connected
The nomadic lifestyle is not right for everyone, but the only way to know for sure is to try. If you can afford to take the risk, go for it. Javier Cuello shares his experience and insights from his four years of travel and work. Read article →
For many years, my own web development company operated as an outsourced web development provider for design agencies. This involved doing everything from front-end development to implementing e-commerce and custom content management solutions. Our direct client was the designer or design agency who had brought us on board to help with the development aspect of the work, however, in an ideal situation, we would be part of the team working to deliver a great end result to the end client.
Sometimes this relationship worked well. We would feel a valued part of the team, our ideas and experience would count, we would work with the designers to come up with the best solution within budgetary, time, and other constraints.
In many cases, however, no attempt was made to form a team. The design agency would throw a picture of a website as a PDF file over the fence to us, then move on to work on their next project. There was little room for collaboration, and often the designer who had created the files was busy on some other work when we came back with questions.
It was an unsatisfactory way to work for everyone. We would be frustrated because we did not have a chance to help ensure that what was designed was possible to be built in a performant and accessible way, within the time and budget agreed on. The designer of the project would be frustrated: Why were these developers asking so many questions? Can they not just build the website as I have designed? Why are the fonts not the size I wanted?
The Waterfall versus Agile argument might be raised here. The situation where a PDF is thrown over the fence is often cited as an example of how bad a Waterfall approach is. Still, working in a fully Agile way is often not possible for teams made of freelancers or separate parties doing different parts of the work. Therefore, in reading these suggestions, look at them through the lens of the projects you work on. However, try not to completely discount something as unworkable because you can’t use the full process. There are often things we can take without needing to fully adopt one methodology or another.
Setting Up A Project For Success
I came to realize that very often the success of failure of the collaboration started before we even won the project, with the way in which we proposed the working relationship. We had to explain upfront that experience had taught us that the approach of us being handed a PDF, quoting and returning a website did not give the best results.
Projects that were successful had a far more iterative approach. It might not be possible to have us work alongside the designers or in a more Agile way. However, having a number of rounds of design and development with time for feedback from each side went a long way to prevent the frustrations of a method where work was completed by each side independently.
Creating Working Relationships
Having longer-term relationships with an agency, spanning a number of projects worked well. We got to know the designers, learned how they worked, could anticipate their questions and ensure that we answered them upfront. We were able to share development knowledge, the things that made a design easier or harder to implement which would, therefore, have an impact on time and budget. They were able to communicate better with us in order to explain why a certain design element was vital, even if it was going to add complexity.
For many freelance designers and developers, and also for those people who work for a distributed company, communication can become mostly text-based. This can make it particularly hard to build relationships. There might be a lot of communication — by email, in Slack, or through messages on a project management platform such as Basecamp. However, all of these methods leave us without the visual cues we might pick up from in-person meetings. An email we see as to the point may come across to the reader as if we are angry. The quick-fire nature of tools such as Slack might leave us committing in writing something which we would not say to that person while looking into their eyes!
Freelance data scientist Nadieh Bremer will talk to us about visualizing data in Toronto. She has learned that meeting people face to face — or at least having a video call — is important. She told me:
“As a remote freelancer, I know that to interact well with my clients I really need to have a video call (stress on the video) I need to see their face and facial/body interactions and they need to see mine. For clients that I have within public transport distance, I used to travel there for a first ‘getting to know each other/see if we can do a project’ meeting, which would take loads of time. But I noticed for my clients abroad (that I can’t visit anyway) that a first client call (again, make sure it’s a video-call) works more than good enough.
It’s the perfect way to weed out the clients that need other skills that I can give, those that are looking for a cheap deal, and those where I just felt something wasn’t quite clicking or I’m not enthusiastic about the project after they’ve given me a better explanation. So these days I also ask my clients in the Netherlands, where I live, that might want to do a first meeting to have it online (and once we get on to an actual contract I can come by if it’s beneficial).”
Working In The Open
Working in the open (with the project frequently deployed to a staging server that everyone had access to see), helped to support an iterative approach to development. I found that it was important to support that live version with explanations and notes of what to look at and test and what was still half finished. If I just invited people to look at it without that information we would get lists of fixes to make to unfinished features, which is a waste of time for the person doing the reporting. However, a live staging version, plus notes in a collaboration tool such as Basecamp meant that we could deploy sections and post asking for feedback on specific things. This helped to keep everyone up to date and part of the project even if — as was often the case for designers in an agency — they had a number of other projects to work on.
There are collaboration tools to help designers to share their work too. Asking for recommendations on Twitter gave me suggestions for Zeplin, Invision, Figma, and Adobe XD. Showing work in progress to a developer can help them to catch things that might be tricky before they are signed off by the client. By sharing the goal behind a particular design feature within the team, a way forward can be devised that meets the goal without blowing the budget.
Scope Creep And Change Requests
The thing about working in the open is that people then start to have ideas (which should be a positive thing), however, most timescales and budgets are not infinite! This means you need to learn to deal with scope creep and change requests in a way that maintains a good working relationship.
We would often get requests for things that were trivial to implement with a message saying how sorry they were about this huge change and requests for incredibly time-consuming things with an assumption it would be quick. Someone who is not a specialist has no idea how long anything will take. Why should they? It is important to remember this rather than getting frustrated about the big changes that are being asked for. Have a conversation about the change, explain why it is more complex than it might appear, and try to work out whether this is a vital addition or change, or just a nice idea that someone has had.
If the change is not essential, then it may be enough to log it somewhere as a phase two request, demonstrating that it has been heard and won’t be forgotten. If the big change is still being requested, we would outline the time it would take and give options. This might mean dropping some other feature if a project has a fixed budget and tight deadline. If there was flexibility then we could outline the implications on both costs and end date.
With regard to costs and timescales, we learned early on to pad our project quotes in order that we could absorb some small changes without needing to increase costs or delay completion. This helped with the relationship between the agency and ourselves as they didn’t feel as if they were being constantly nickel and dimed. Small changes were expected as part of the process of development. I also never wrote these up in a quote as contingency, as a client would read that and think they should be able to get the project done without dipping into the contingency. I just added the time to the quote for the overall project. If the project ran smoothly and we didn’t need that time and money, then the client got a smaller bill. No one is ever unhappy about being invoiced for less than they expected!
This approach can work even for people working in-house. Adding some time to your estimates means that you can absorb small changes without needing to extend the timescales. It helps working relationships if you are someone who is able to say yes as often as possible.
This does require that you become adept at estimating timescales. This is a skill you can develop by logging your time to achieve your work, even if you don’t need to log your time for work purposes. While many of the things you design or develop will be unique, and seem impossible to estimate, by consistently logging your time you will generally find that your ballpark estimates become more accurate as you make yourself aware of how long things really take.
“It all comes down to respect for your colleague’s craft, and sort of knowing your place and precisely where you fit into the project. When working with a developer, I surrender to them in a creative way, and then, defuse whatever power play they might try to make on me by leading the charges with constructive design advice, lightning-fast email replies and generally keeping the spirit upbeat. It’s an odd offense to play. I’m not down with the adversarial stuff. I’m quick to remind them we are all in the same boat, and, who’s paying their paycheck. And that’s not me. It’s the client. I’ll forever be on their team, you know? We make the stuff for the client. Not just me. Not ‘my team’. We do it together. This simple methodology has always gone a long way for me.”
I love this, it underpins everything that this article discusses. Think back to any working relationship that has gone bad, how many of those involved you feeling as if the other person just didn’t understand your point of view or the things you believe are important? Most reasonable people understand that compromise has to be made, it is when it appears that your point of view is not considered that frustration sets in.
There are sometimes situations where a decision is being made, and your experience tells you it is going to result in a bad outcome for the project, yet you are overruled. On a few occasions, decisions were made that I believed so poor; I asked for the decision and our objection to it be put in writing, in order that we could not be held accountable for any bad outcome in future. This is not something you should feel the need to do often, however, it is quite powerful and sometimes results in the decision being reversed. An example would be of a client who keeps insisting on doing something that would cause an accessibility problem for a section of their potential audience. If explaining the issue does not help, and the client insists on continuing, ask for that decision in writing in order to document your professional advice.
Learning The Language
I recently had the chance to bring my CSS Layout Workshop not to my usual groups of front-end developers but instead to a group of UX designers. Many of the attendees were there not to improve their front-end development skills, but more to understand enough of how modern CSS Layout worked that they could have better conversations with the developers who built their designs. Many of them had also spent years being told that certain things were not possible on the web, but were realizing that the possibilities in CSS were changing through things like CSS Grid. They were learning some CSS not necessarily to become proficient in shipping it to production, but so they could share a common language with developers.
There are often debates on whether “designers should learn to code.” In reality, I think we all need to learn something of the language, skills, and priorities of the other people on our teams. As Aaron reminded us, we are all on the same team, we are making stuff together. Designers should learn something about code just as developers should also learn something of design. This gives us more of a shared language and understanding.
“I have basically made a career out of being both technical and creative so I strongly feel that the more crossover the better. Obviously what I do now is wonderfully free of the constraints of client work but even so, I do think that if you can blur those edges, it’s gonna be good for you. It’s why I speak at design conferences and encourage designers to play with creative coding, and I speak at tech conferences to persuade coders to improve their visual acuity. Also with creative coding. It’s good because not only do I get to work across both disciplines, but also I get to annoy both designers and coders in equal measure.”
I have found that introducing designers to browser DevTools (in particular the layout tools in Firefox and also to various code generators on the web) has been helpful. By being able to test ideas out without writing code, helps a designer who isn’t confident in writing code to have better conversations with their developer colleagues. Playing with tools such as gradient generators, clip-path or animation tools can also help designers see what is possible on the web today.
We are also seeing a number of tools that can help people create websites in a more visual way. Developers can sometimes turn their noses up about the code output of such tools, and it’s true they probably won’t be the best choice for the production code of a large project. However, they can be an excellent way for everyone to prototype ideas, without needing to write code. Those prototypes can then be turned into robust, permanent and scalable versions for production.
An important tip for developers is to refrain from commenting on the code quality of prototypes from members of the team who do not ship production code! Stick to what the prototype is showing as opposed to how it has been built.
A Practical Suggestion To Make Things Visual
Eva-Lotta Lamm will be speaking in Toronto about Sketching and perhaps unsurprisingly passed on practical tips for helping conversation by visualizing the problem to support a conversation.
Creating a shared picture of a problem or a solution is a simple but powerful tool to create understanding and make sure they everybody is talking about the same thing.
Visualizing a problem can reach from quick sketches on a whiteboard to more complex diagrams, like customer journey diagrams or service blueprints.
But even just spatially distributing words on a surface adds a valuable layer of meaning. Something as simple as arranging post-its on a whiteboard in different ways can help us to see relationships, notice patterns, find gaps and spot outliers or anomalies. If we add simple structural elements (like arrows, connectors, frames, and dividers) and some sketches into the mix, the relationships become even more obvious.
Visualising a problem creates context and builds a structural frame that future information, questions, and ideas can be added to in a ‘systematic’ way.
Visuals are great to support a conversation, especially when the conversation is ‘messy’ and several people involved.
When we visualize a conversation, we create an external memory of the content, that is visible to everybody and that can easily be referred back to. We don’t have to hold everything in our mind. This frees up space in everybody’s mind to think and talk about other things without the fear of forgetting something important. Visuals also give us something concrete to hold on to and to follow along while listening to complex or abstract information.
When we have a visual map, we can point to particular pieces of content — a simple but powerful way to make sure everybody is talking about the same thing. And when referring back to something discussed earlier, the map automatically reminds us of the context and the connections to surrounding topics.
When we sketch out a problem, a solution or an idea the way we see it (literally) changes. Every time we express a thought in a different medium, we are forced to shape it in a specific way, which allows us to observe and analyze it from different angles.
Visualising forces us to make decisions about a problem that words alone don’t. We have to decide where to place each element, decide on its shape, size, its boldness, and color. We have to decide what we sketch and what we write. All these decisions require a deeper understanding of the problem and make important questions surface fairly quickly.
All in all, supporting your collaboration by making it more visual works like a catalyst for faster and better understanding.
Working in this way is obviously easier if your team is working in the same room. For distributed teams and freelancers, there are alternatives to communicate in ways other than words, e.g. by making a quick Screencast to demonstrate an issue, or even sketching and photographing a diagram can be incredibly helpful. There are collaborative tools such as Milanote, Mural, and Niice; such tools can help with the process Eva-Lotta described even if people can’t be in the same room.
I’m very non-visual and have had to learn how useful these other methods of communication are to the people I work with. I have been guilty on many occasions of forgetting that just because I don’t personally find something useful, it is still helpful to other people. It is certainly a good idea to change how you are trying to communicate an idea if it becomes obvious that you are talking at cross-purposes.
Over To You
As with most things, there are many ways to work together. Even for remote teams, there is a range of tools which can help break down barriers to collaborating in a more visual way. However, no tool is able to fix problems caused by a lack of respect for the work of the rest of the team. A good relationship starts with the ability for all of us to take a step back from our strongly held opinions, listen to our colleagues, and learn to compromise. We can then choose tools and workflows which help to support that understanding that we are all on the same team, all trying to do a great job, and all have important viewpoints and experience to bring to the project.
I would love to hear your own experiences working together in the same room or remotely. What has worked well — or not worked at all! Tools, techniques, and lessons learned are all welcome in the comments. If you would be keen to see tutorials about specific tools or workflows mentioned here, perhaps add a suggestion to our User Suggestions board, too.
On Failures And Successes: Meet SmashingConf Freiburg 2018
Everybody loves speaking about successes, but nobody can succeed without failing big time along the way. It’s through mistakes that we grow and get smarter. So for the upcoming SmashingConf Freiburg 2018 (Sept. 10–11), we want to put these stories into focus for a change and explore practical techniques and strategies learned in real projects — the hard way. Aarron Walter, Josh Clark, Tammy Everts, Morten Rand-Hendriksen & many others. Sept 10–11. Early-Birds are available now →
The night before the conference we’ll be hosting a FailNight — a warm-up party with a twist. Every session will be highlighting how we all failed on a small or big scale, and what we all can learn from it. With talks from the community, for the community. Sounds like fun? Well, it will be!
As usual, one track, two conference days (Sept. 10–11), 12 speakers, and just 260 available seats. The conference will cover everything from efficient design workflow to design systems and copywriting, multi-cultural designs, designing for mobile and other fields that may come up in your day-to-day work.
First confirmed speakers include:
Design for Machine Learning Josh Clark(Big Medium)
Our workshops give you the opportunity to spend a full day on the topic of your choice. Tickets for the full-day workshops cost €399. If you buy a workshop ticket in combination with a conference ticket, you’ll save €100 on the regular workshop ticket price. Seats are limited
Workshops on Wednesday, September 12th
Josh Clark on Design For What’s Next Spend a day exploring the web’s emerging interactions and how you can put them to work today. Your guide is designer Josh Clark, author of Designing for Touch and ambassador of the near future. As you move into newer design tools — speech, bots, physical interfaces, artificial intelligence, and more — you’ll learn the tools and techniques for prototyping and launching these new interfaces and get answers to foundational questions for all your projects. Read more…
Vitaly Friedman on Dirty Little Tricks From The Dark Corners Of eCommerce In this workshop, Vitaly will use real-life examples as a case study and examine refinements of the interface on spot. You’ll set up a very clear roadmap on how you can do the right things in the right order to improve conversion and customer experience. That means removing distractions, minimizing friction and avoiding disruptions and dead ends caused by the interface. Read more…
As always, the Historical Merchants’ Hall located right in the heart of our hometown Freiburg will be the home of SmashingConf Freiburg. First mentioned in 1378 and having retained its present-day form since 1520, the “Kaufhaus” is a symbol of the importance of trade in medieval Freiburg, and, well, its beautiful architecture still blows our audience away each year anew.
Why This Conference Could Be For You
Each SmashingConf is a friendly and intimate experience. A cozy get-together of likeminded people who share their stories, their ideas, their hard-learned lessons. At SmashingConf Freiburg you will learn how to:
Use production-ready CSS Grid layouts,
Recognize, revise, and resolve dark patterns and misleading copy in your own products,
Design and build a product with a global audience in mind,
Extract action-oriented insights from real user data,
Create better e-commerce experiences,
Create responsible machine-learning applications,
Get leading design right,
… and a lot more.
Download “Convince Your Boss” PDF
You need to convince your boss to send you to Freiburg? No worries, we’ve prepared a neat Convince Your Boss PDF that you can use to tip the scales in your favor. Fingers crossed.
Redesigning A Digital Interior Design Shop (A Case Study)
Good products are the result of a continual effort in research and design. And, as it usually turns out, our designs don’t solve the problems they were meant to right away. It’s always about constant improvement and iteration.
I have a client called Design Cafe (let’s call it DC). It’s an innovative interior design shop founded by a couple of very talented architects. They produce bespoke designs for the Indian market and sell them online.
DC approached me two years ago to design a few visual mockups for their website. My scope then was limited to visuals, but I didn’t have the proper foundation upon which to base those visuals, and since I didn’t have an ongoing collaboration with the development team, the final website design did not accurately capture the original design intent and did not meet all of the key user needs.
A year and a half passed and DC decided to come back to me. Their website wasn’t providing the anticipated stream of leads. They came back because my process was good, but they wanted to expand the scope to give it space to scale. This time, I was hired to do the research, planning, visual design and prototyping. This would be a makeover of the old design based on user input and data, and prototyping would allow for easy communication with the development team. I assembled a small team of two: me and a fellow designer, Miroslav Kirov, to help run proper research. In less than two weeks, we were ready to start.
Useful tip: I always kick off a project by talking to the stakeholders. For smaller projects with one or two stakeholders, you can blend the kick-off and the interview into one. Just make sure it’s no longer than an hour.
Our two stakeholders are both domain experts. They have a brick-and-mortar store in the center of Bangalore that attracts a lot of people. Once in there, people are delighted by the way the designs look and feel. Our clients wanted to have a website that conveys the same feeling online and that would make its visitors want to go to the store.
There wasn’t a clear distinction between new, returning and potential clients.
DC’s selling points weren’t clearly communicated.
They had future plans for transforming the website into a hub for interior design ideas. And, last but not least, DC wanted to attract fresh design talent.
Defining the Goals
We shortlisted all of our goals for the project. Our main goal was to explain in a clear and appealing manner what DC does for existing and potential clients in a way that engages them to contact DC and go to the store. Some secondary goals were:
lower the drop-off rate,
capture some customer data,
clarify the brand’s message,
make the website responsive,
explain budgets better,
provide decision-making assistance and become an information influencer.
Our number-one key metric was to convert users to leads who visit the store, which measures the main goal. We needed to improve that by at least 5% initially — a realistic number we decided on with our stakeholders. In order to do that, we needed to:
shorten the conversion time (time needed for a user to get in touch with DC),
increase the form application rate,
increase the overall satisfaction users get from the website.
We would track these metrics by setting up Google Analytics Events once the website is online and by talking with leads who come into the store through the website.
Useful tip: Don’t focus on too many metrics. A handful of your most important ones are enough. Measuring too many things will dilute the results.
In order for us to gain the best possible insights, our user interviews had to target both previous and potential clients, but we had to go minimal, so we picked two potential and three existing clients. They were mostly from the IT sector — DC’s main target group. Given our pretty tight schedule, we started with desk research while we waited for all five user interviews to be scheduled.
Useful tip: You need to know who you are designing for and what research has been done before. Stakeholders tell you their story, but you need to compare it to data and to users’ opinions, expectations and needs.
We could reference some Google Analytics data from the website:
Most users went to the kitchen, then to the bedroom, then to the living room.
The high bounce rate of 80%+ was probably due to a misunderstanding of the brand message and unclear flows and calls to action (CTAs).
Traffic was mostly mobile.
Most users landed on the home page, 70% of them from ads and 16% directly (mostly returning customers), and the rest were equally divided between Facebook and Google Search.
90% of social media traffic came from Facebook. Expanding brand awareness to Instagram and Twitter could be beneficial.
There’s a lot of local competition in the sector. Here were some repeating patterns:
video spots and elaborate galleries showing the completed designs with clients discussing their services;
attractive design presentations with high-quality photos;
targeting of group’s appropriate messages;
quizzes for picking styles;
big bold typography, less text and more visuals.
DC’s customers are mostly aged between 28 and 40, with a secondary set in the higher bracket of 38 and 55 who come for their second home. They are IT or business professionals with a mid to high budget. They value good customer experience but are price-conscious and very practical. Because they are mostly families, very often the wives are the hidden dominant decision-maker.
We talked with five users (three existing and two potential customers) and sent out a survey to 20 more (mixing existing and potential customers; see Design Cafe Questionnaire).
Useful tip: Be sure to schedule all of your interviews ahead of time, and plan for more people than you need. Include extreme users along with the mainstreams. Chances are that if something works for an extreme user, it will work for the rest as well. Extremes will also give you insight about edge cases that mainstreams just don’t care about.
All users were confused about the main goal of the website. Some of their opinions:
“It lacks a proper flow.”
“I need more clarity in the process, especially in terms of timelines.”
“I need more educational information about interior design.”
Everyone was pretty well informed about the competition. They had tried other companies before DC. All found out about DC by either a reference, Google, ads or by physically passing by the store. And, boy, did they love the store! They treated it like an Apple Store for interior design. Turns out that DC really did a great job with that.
Useful tip: Negative feedback helps us find opportunities for improvement. But positive feedback is also pretty useful because it helps you identify which parts of the product are worth retaining and building upon.
Personal touch, customer service, prices and quality of materials were their main motivations for choosing DC. People insisted on being able to see the price of every element on a page at any time (the previous design didn’t have prices on the accessories).
We made an interesting but somehow expected discovery about device usage. Mobile devices were used mostly for consumption and browsing, but when it came to ordering, most people opened their laptops.
The survey results mostly overlapped with the interviews:
Users found DC through different channels, but mainly through referrals.
They didn’t quite understand the current state of the website. Most of them had searched for or used other services before DC.
All of the surveyed users ordered kitchen designs. Almost all had difficulty choosing the right design style.
Most users found the process of designing their own interior hard and were interested in features that could make their choice easier.
Useful tip: Writing good survey questions takes time. Work with a researcher to write them, and schedule double the time you think you’ll need.
User Journeys Overview
Talking with customers helped us gain useful insight about which scenarios would be most important to them. We made an affinity diagram with everything we collected and started prioritizing and combining items in chunks.
The result was seven point-of-view problem statements that we decided to design for:
A new customer needs more information about DC because they need proof of credibility.
A returning customer needs quick access to the designs because they don’t want to waste time.
All customers need to be able to browse the designs at any time.
All customers want to browse designs relevant to their tastes, because that will shorten their search time.
Potential leads need a way to get in touch with DC in order to purchase a design.
All customers, once they’ve ordered, need to stay up to date with their order status, because they need to know what they are paying for and when they will be getting it.
All customers want to read case studies about successful projects, because that will reassure them that DC knows its stuff.
Using this list, we came up with design solutions for every journey.
The previous home page of Design Cafe was confusing. It needed to present more information about the business. The lack of information caused confusion and people were unsure what DC is about. We divided the home page into several sections and designed it so that every section could satisfy the needs of one of our target groups:
For new visitors (the purple flow), we included a short trip through the main unique selling points (USPs) of the service, the way it works, some success stories and an option to start the style quiz.
For returning visitors (the blue flow), who will most likely skip the home page or use it as a waypoint, the hero section and the navigation pointed a way out to browsing designs.
We left a small part at the end of the page (the orange flow) for potential employees, describing what there is to love about DC and a CTA that goes to the careers page.
The whole point of the onboarding process was to capture the customer’s attention so that they could continue forward, either directly to the design catalog or through a feature we called the style quiz.
We made the style quiz to help users narrow down their results.
DC previously had a feature called a 3D builder that we decided to remove. It allowed you to set your room size and then drag-and-drop furniture, windows and doors into the mix. In theory, this sounds good, but in reality people treated it much like a game and expected it to function like a minified version of The Sims’ Build Mode.
Everything made with the 3D builder was ending up completely modified by the designers. The tool was giving people a lot of design power and too many choices. On top of that, supporting it was a huge technical endeavor because it was a whole product on its own.
Compared to it, the style quiz was a relatively simple feature:
It starts out by asking about colors, textures and designs you like.
It continues to ask about room type.
Eventually, it displays a curated list of designs based on your answers.
The whole quiz wizard extends to only four steps and takes less than a minute to complete. But it makes people invest a tad bit of their time, thus creating engagement. The result: We’re improving conversion time and overall satisfaction.
Alternatively, users can skip the style quiz and go directly to the design catalog, then use the filters to fine-tune the results. The page automatically shows kitchen designs, what most people are looking for. And for the price-conscious, we made a small feature that allows them to input their room’s size, and all prices are recalculated.
If people don’t like anything from the catalog, chances are they are not DC’s target customer and there’s not much we can do to keep them on the website. But if they do like a design, they could decide to go forward and get in touch with DC, which brings us to the next step in the process.
Getting in Touch
Contacting DC needed to be as simple as possible. We implemented three ways to do that:
through the chat, shown on every page — the quickest way;
by opening the contact page and filling out the form or by just calling DC on the phone;
by clicking “Book a consultation” in the header, which asks for basic information and requests an appointment (upon submission, the next steps are shown to let users know what exactly is going to happen).
The rest of this journey continues offline: Potential customers meet a DC designer and, after some discussions and planning, place an order. DC notifies them of any progress via email and sends them a link to the progress tracker.
The progress tracker is in a user menu in the top-right corner of the design. Its goal is to show a timeline of the order. Upon an update, an “unread” notification pops out. Most users, however, will usually find out about order updates through email, so the entry point for the whole flow will be external.
Once the interior design order is installed and ready, users will have the completed order on the website for future reference. Their project could be featured on the home page and become part of the case studies.
One of DC’s long-term goals is for its website to become an influencer hub for interior design, filled with case studies, advice and tips. It’s part of a commitment to providing quality content. But DC doesn’t have that content yet. So, we decided to start that section with minimal effort and introduce it as a blog. The client would gradually fill it up with content and detailed process walkthroughs. These would be later expanded and featured on the home page. Case studies are a feature that could significantly increase brand awareness, though they would take time.
Preparing for Visual Design
With the critical user journeys all figured out and wireframed, we were ready to delve into visual design.
Data showed that most people open the website on their phones, but interviews proved that most of them were more willing to buy through a computer, rather than a mobile device. Also, desktop and laptop users were more engaged and loyal. So, we decided to design for desktop-first and work down to the smaller (mobile) resolutions from it in code.
We started collecting visual ideas, words and images. Initially, we had a simple word sequence based on our conversations with the client and a mood board with relevant designs and ideas. The main visual features we were after were simplicity, bold typography, nice photos and clean icons.
Useful tip: Don’t follow a certain trend just because everybody else is doing it. Create a thorough mood board of relevant reference designs that approximate the look and feel you’re going after. This look should be in line with your goals and target audience.
Our client had already started working on a photo shoot, and the results were great. Stock photography would have ruined everything personal about this website. The resulting photos blended with the big type pretty well and helped with that simple language we were after.
Initially, we went with a combination of Raleway and Roboto for the typography. Raleway is a great font but a bit overused. The second iteration was Abril Fatface and Raleway for the copy. Abril Fatface resembles the splendor of Didot and made the whole page a lot more heavy and pretentious. It was an interesting direction to explore, but it didn’t resonate with the modern techy feel of DC. The last iteration was Nexa for the titles, which turned out to be the best choice due to its modern and edgy feel, with Lato — both a great fit.
Useful tip: Play around with type variations. List them side by side to see how they compare. Go to Typewolf, MyFonts or a similar website to get inspired. Look for typefaces that make sense for your product. Consider readability and accessibility. Don’t go overboard with your type scale; keep it as minimal as possible. Check out Butterick’s summary of key rules if in doubt.
DC already had a color scheme, but they gave us the freedom to experiment. The main colors were tints of cyan, golden and plum (or, rather, a strange kind of bordeaux), but the original hues were too faded and didn’t blend with each other well enough.
Useful tip: If the brand already has colors, test slight variations to see how they fit the overall design. Or remove some of the colors and use only one or two. Try designing your layout in monochrome and then test different color combinations on an already mocked-up design. Check out some other great tips by Wojciech Zieliński in his article “How to Use Colors in UI Design: Practical Tips and Tools”.
Here’s what we decided on in the end:
The way we presented all of those type variants and colors was through iterations on the home page.
We focused the first visual iteration on getting the main information clearly visible and squeezing the most out of the testimonials and style quiz sections. After some discussion, we figured it was too plain and needed improvement. We made changes to the fonts and icons and modified some sections, shown in iterations 2 and 3 in the image below.
We didn’t have the time to design custom icons, but the NounProject came to the rescue. With the SVG file format, it’s very simple to change whatever you need and mix it with something else. This sped up our work immensely, and with visual iteration number 4, we signed off on the design of the home page. This allowed us to focus on components and use them as LEGO blocks to build the templates.
I listed most components (see PDF) in a Sketch artboard to keep them accessible. Whenever the design needed a new pattern, we’d come back to this page and look for ways to reuse elements. Having a visual system in place, even for a small project like this, kept things consistent and simple.
Useful tip: Components, atoms, blocks — no matter what you call them, they are all part of systematic thinking about your design. Design systems help you gain a deeper understanding of your product by urging you to focus on patterns, design principles and design language. If you’re new to this approach, check out Brad Frost’s Atomic Design or Alla Kholmatova’s Design Systems.
Prototyping With Code
For our prototype, we decided to use code and set up a simple build process to speed up our work.
Picking tools and processes
Gulp automated everything. If you haven’t heard of it, check out Callum Macrae’s awesome guide. Gulp enabled us to handle all of the styles, scripts and templates, and it outputs a ready-to-use minified production version of the code.
Some of the more important Gulp plugins we used were:
This allows you to use PostCSS. You can bundle it with plugins like cssnext to get a pretty robust and versatile setup.
This sets up a server and automatically updates the view on every change. You can set it to fire up upon starting “gulp watch”, and everything will be synced up on hitting “Save”.
This is a Handlebars implementation for Gulp. It’s a quick way to create templates and reuse them. Imagine you have a button that stays the same throughout the whole design. It would be a symbol in Sketch. It’s basically the same concept but wrapped in HTML. Whenever you want to use that button, you just include the button template. If you change something in the master template, it propagates the changes to every other button in the design. You do that for everything in the design system, and thus you’re using the same paradigm for both visual design and code. No more static page mockups!
Components and templates
We had to mix atomic CSS with module-based CSS to get the most of both worlds. Atomic CSS handled all of the general styles, while the CSS modules handled edge cases.
In atomic CSS, atoms are immutable CSS classes that do just one thing. We used Tachyons, an atomic toolkit. In Tachyons, every class you apply is a single CSS property. For instance, .b stands for font-weight: bold, and .ttu stands for text-transform: uppercase. A paragraph with bold uppercase text would look like this:
<p class="b ttu">Paragraph</p>
Useful tip: Once you get familiar with atomic CSS, it becomes a blazingly fast way to prototype stuff — and a very systematic one, because it urges you to constantly think about reusability and optimization.
A major benefit of prototyping with code is that you can demo complex interactions. We coded most of our critical journeys this way.
Useful tip: With HTML prototypes, you will have to decide the level of fidelity you want to achieve. That might get pretty time-consuming if you go too deep. But you can’t really go wrong with that either because as you go deeper and deeper into the code and fine-tune every possible detail, at some point you’ll start delivering the actual product.
Clients, especially small B2C companies, love when you deliver a design solution that they can use immediately. We shipped just that.
Unfortunately, you can’t always predict a project’s pace, and it took several months for our code to be integrated in DC’s workflow. In its current state, this code is ready for testing, and what’s better is that it’s pretty easy to modify. So, if DC decides to conduct some user tests in the future, any changes will be easy to make.
Collaborate with other designers whenever possible. When two people are thinking about the same problem, they will deliver better ideas. Take turns in taking notes during interviews, and brainstorm goals, ideas and visuals together.
We shipped a working version of the website, and the client was able to use it right away. If you aren’t able to sign off on the code, try to get as close to the final product as possible, and communicate that visually to your client’s team. Document your design — it’s a deliverable that will be used and abused by everyone, from developers to marketers to in-house designers. Set aside some time to make sure all of your ideas are properly understood by everyone.
Scheduling interviews and writing good surveys can be time-consuming. You have to plan ahead and recruit more people than you think you will need. Hire an experienced researcher to work with you on these tasks, and spend some time with your team to identify your goals. Be careful when sourcing participants. Your client can help you find the right people, but you’ll need to stick to participants who meet the right demographics.
Schedule enough time for planning. Project goals, processes, and responsibilities should be clear to everyone on your team. You need time to allow for multiple iterations on prototypes, because prototypes improve products quickly. If you don’t want to mess with code, there are various ways to prototype. But even if you do, you don’t need to write flawless code — just write designer’s code. Or, as Alan Cooper once said, “Sometimes the best way for a designer to communicate their vision is to code something up so that their colleagues can interact with the proposed behavior, rather than just see still images. The goal of such code is not the same as the goal of the code that coders write. The code isn’t for deployment, but for design [and] its purpose is different.”
Don’t focus on a unique design per se, unless that’s the main feature of your product. Better to spend time on things that matter more. Use frameworks, icons and visual assets where possible, or outsource them to another designer and focus on your core product goals and metrics.
What You Need To Know To Increase Mobile Checkout Conversions
Google’s mobile-first indexing is here. Well, for some websites anyway. For the rest of us, it will be here soon enough, and our websites need to be in tip-top shape if we don’t want search rankings to be adversely affected by the change.
That said, responsive web design is nothing new. We’ve been creating custom mobile user experiences for years now, so most of our websites should be well poised to take this on… right?
Here’s the problem: Research shows that the dominant device through which users access the web, on average, is the smartphone. Granted, this might not be the case for every website, but the data indicates that this is the direction we’re headed in, and so every web designer should be prepared for it.
However, mobile checkout conversions are, to put it bluntly, not good. There are a number of reasons for this, but that doesn’t mean that m-commerce designers should take this lying down.
As more mobile users rely on their smart devices to access the web, websites need to be more adeptly designed to give them the simplified, convenient and secure checkout experience they want. In the following roundup, I’m going to explore some of the impediments to conversion in the mobile checkout and focus on what web designers can do to improve the experience.
Why Are Mobile Checkout Conversions Lagging?
According to the data, prioritizing the mobile experience in our web design strategies is a smart move for everyone involved. With people spending roughly 51% of their time with digital media through mobile devices (as opposed to only 42% on desktop), search engines and websites really do need to align with user trends.
Now, while that statistic paints a positive picture in support of designing websites with a mobile-first approach, other statistics are floating around that might make you wary of it. Here’s why I say that: Monetate’s e-commerce quarterly report issued for Q1 2017 had some really interesting data to show.
In this first table, they break down the percentage of visitors to e-commerce websites using different devices between Q1 2016 and Q1 2017. As you can see, smartphone Internet access has indeed surpassed desktop:
Website Visits by Device
Monetate’s findings on which devices are used to access in the Internet. (Source)
In this next data set, we can see that the average conversion rate for e-commerce websites isn’t great. In fact, the number has gone down significantly since the first quarter of 2016.
Monetate’s findings on overall e-commerce global conversion rates (for all devices). (Source)
Even more shocking is the split between device conversion rates:
Conversion Rates by Device
Monetate’s findings on the average conversion rates, broken down by device. (Source)
Smartphones consistently receive fewer conversions than desktop, despite being the predominant device through which users access the web.
What’s the problem here? Why are we able to get people to mobile websites, but we lose them at checkout?
In its report from 2017 named “Mobile’s Hierarchy of Needs,” comScore breaks down the top five reasons why mobile checkout conversion rates are so low:
Here is the breakdown for why mobile users don’t convert:
20.2% — security concerns
19.6% — unclear product details
19.6% — inability to open multiple browser tabs to compare
19.3% — difficulty navigating
18.6% — difficulty inputting information.
Those are plausible reasons to move from the smartphone to the desktop to complete a purchase (if they haven’t been completely turned off by the experience by that point, that is).
In sum, we know that consumers want to access the web through their mobile devices. We also know that barriers to conversion are keeping them from staying put. So, how do we deal with this?
10 Ways to Increase Mobile Checkout Conversions In 2018
For most of the websites you’ve designed, you’re not likely to see much of a change in search ranking when Google’s mobile-first indexing becomes official.
Your mobile-friendly designs might be “good enough” to keep your websites at the top of search (to start, anyway), but what happens if visitors don’t stick around to convert? Will Google start penalizing you because your website can’t seal the deal with the majority of visitors? In all honesty, that scenario will only occur in extreme cases, where the mobile checkout is so poorly constructed that bounce rates skyrocket and people stop wanting to visit the website at all.
Let’s say that the drop-off in traffic at checkout doesn’t incur penalties from Google. That’s great… for SEO purposes. But what about for business? Your goal is to get visitors to convert without distraction and without friction. Yet, that seems to be what mobile visitors get.
Going forward, your goal needs to be two-fold:
to design websites with Google’s mobile-first mission and guidelines in mind,
to keep mobile users on the website until they complete a purchase.
Essentially, this means decreasing the amount of work users have to do and improving the visibility of your security measures. Here is what you can do to more effectively design mobile checkouts for conversions.
1. Keep the Essentials in the Thumb Zone
Research on how users hold their mobile phones is old hat by now. We know that, whether they use the single- or double-handed approach, certain parts of the mobile screen are just inconvenient for mobile users to reach. And when expediency is expected during checkout, this is something you don’t want to mess around with.
For single-handed users, the middle of the screen is the prime playing field:
Although users who cradle their phones for greater stability have a couple options for which fingers to use to interact with the screen, only 28% use their index finger. So, let’s focus on the capabilities of thumb users, which, again, means giving the central part of the screen the most prominence:
Some users hold their phones with two hands. Because the horizontal orientation is more likely to be used for video, this won’t be relevant for mobile checkout. So, pay attention to how much space of that screen is feasibly within reach of the user’s thumb:
In sum, we can use Smashing Magazine’s breakdown of where to focus content, regardless of left-hand, right-hand or two-handed holding of a smartphone:
While information is included at the top of the checkout page, the input fields don’t start until just below the middle of it — directly in the ideal thumb zone for users of any type. This ensures that visitors holding their phones in any manner and using different fingers to engage with it will have no issue reaching the form fields.
2. Minimize Content to Maximize Speed
We’ve been taught over and over again that minimal design is best for websites. This is especially true in mobile checkout, where an already slow or frustrating experience could easily push a customer over the edge, when all they want to do is be done with the purchase.
To maximize speed during the mobile checkout process, keep the following tips in mind:
Only add the essentials to checkout. This is not the time to try to upsell or cross-sell, promote social media or otherwise distract from the action at hand.
Keep the checkout free of all images. The only eye-catching visuals that are really acceptable are trustmarks and calls to action (more on these below).
Any text included on the page should be instructional or descriptive in nature.
Avoid any special stylization of fonts. The less “wow” your checkout page has, the easier it will be for users to get through the process.
Look to Staples’ website as an example of what a highly simple single-page checkout should look like:
As you can see, Staples doesn’t bog down the checkout process with product images, branding, navigation, internal links or anything else that might (1) distract from the task at hand, or (2) suck resources from the server while it attempts to process your customers’ requests.
Not only will this checkout page be easy to get through, but it will load quickly and without issue every time — something customers will remember the next time they need to make a purchase. By keeping your checkout pages light in design, you ensure a speedy experience in all aspects.
3. Put Them at Ease With Trustmarks
A trustmark is any indicator on a website that lets customers know, “Hey, there’s absolutely nothing to worry about here. We’re keeping your information safe!”
The one trustmark that every m-commerce website should have? An SSL certificate. Without one, the address bar will not display the lock sign or the green https domain name — both of which let customers know that the website has extra encryption.
You can use other trustmarks at checkout as well.
While you can use logos from Norton Security, PCI compliance and other security software to let customers know your website is protected, users might also be swayed by recognizable and well-trusted names. When you think about it, this isn’t much different than displaying corporate logos beside customer testimonials or in callouts that boast of your big-name connections. If you can leverage a partnership like the ones mentioned below, you can use the inherent trust there to your benefit.
Take 6pm, which uses a “Login with Amazon” option at checkout:
This is a smart move for a brand that most definitely does not have the brand-name recognition that a company like Amazon has. By giving customers a convenient option to log in with a brand that’s synonymous with speed, reliability and trust, the company might now become known for those same checkout qualities that Amazon is celebrated for.
Then, there are mobile checkout pages like the one on Sephora:
Sephora also uses this technique of leveraging another brand’s good name in order to build trust at checkout time. In this case, however, it presents customers with two clear options: Check out with us right now, or hop over to PayPal, which will take care of you securely. With security being a major concern that keeps mobile customers from converting, this kind of trustmark and payment method is a good move on Sephora’s part.
4. Provide Easier Editing
In general, never take a visitor (on any device) away from whatever they’re doing on your website. There are already enough distractions online; the last thing they need is for you to point them in a direction that keeps them from converting.
At checkout, however, your customers might feel compelled to do this very thing if they decide they want a different color, size or quantity of an item in their shopping cart. Rather than let them backtrack through the website, give them an in-checkout editing option to keep them in place.
When they first get to the checkout screen, customers will see a list of items they’re about to purchase. When the large “Edit” button beside each item is clicked, a lightbox (shown above) opens with the product’s variations. It’s basically the original product page, just superimposed on top of the checkout. Users can adjust their options and save their changes without ever having to leave the checkout page.
If you find, in reviewing your website’s analytics, that users occasionally backtrack after hitting the checkout (you can see this in the sales funnel), add this built-in editing feature. By preventing this unnecessary movement backwards, you could save yourself lost conversions from confused or distracted customers.
5. Enable Express Checkout Options
When consumers check out on an e-commerce website through a desktop device, it probably isn’t a big deal if they have to input their user name, email address or payment information each time. Sure, if it can be avoided, they’ll find ways around it (like allowing the website to save their information or using a password manager such as LastPass).
But on mobile, re-entering that information is a pain, especially if contact forms aren’t optimized well (more on that below). So, to ease the log-in and checkout process for mobile users, consider ways in which you can simplify the process:
Allow for guest checkout.
Allow for one-click expedited checkout.
Enable one-click sign-in from a trusted source, like Facebook.
Enable payment on a trusted payment provider’s website, like PayPal, Google Wallet or Stripe.
One of the nice things about Sephora‘s already convenient checkout process is that customers can automate the sign-in process going forward with a simple toggle:
When mobile customers are feeling the rush and want to get to the next stage of checkout, Sephora’s auto-sign-in feature would definitely come in handy and encourage customers to buy more frequently from the mobile website.
Many mobile websites wait until the bottom of the login page to tell customers what kinds of options they have for checking out. But rather than surprise them late, Victoria’s Secret displays this information in big bold buttons right at the very top:
Customers have a choice of signing in with their account, checking out as a guest or going directly to PayPal. They are not surprised to discover later on that their preferred checkout or payment method isn’t offered.
I also really love how Victoria’s Secret has chosen to do this. There’s something nice about the brightly colored “Sign In” button sitting beside the more muted “Check Out as a Guest” button. For one, it adds a hint of Victoria’s Secret brand colors to the checkout, which is always a nice touch. But the way it’s colored the buttons also makes clear what it wants the primary action to be (i.e. to create an account and sign in).
6. Add Breadcrumbs
When you send mobile customers to checkout, the last thing you want is to give them unnecessary distractions. That’s why the website’s standard navigation bar (or hamburger menu) is typically removed from this page.
Nonetheless, the checkout process can be intimidating if customers don’t know what’s ahead. How many forms will they need to fill out? What sort of information is needed? Will they have a chance to review their order before submitting payment details?
If you’ve designed a multi-page checkout, allay your customers’ fears by defining each step with clearly labeled breadcrumb navigation at the top of the page. In addition, this will give your checkout a cleaner design, reducing the number of clicks and scrolling per page.
Hayneedle has a beautiful example of breadcrumb navigation in action:
You can see that three steps are broken out and clearly labeled. There’s absolutely no question here about what users will encounter in those steps either, which will help put their minds at ease. Three steps seems reasonable enough, and users will have a chance to review the order once more before completing the purchase.
Sephora has an alternative style of “breadcrumbs” in its checkout:
Instead of placing each “breadcrumb” at the top of the checkout page, Sephora’s customers can see what the next step is, as well as how many more are to come as they work their way through the form.
This is a good option to take if you’d rather not make the top navigation or the breadcrumbs sticky. Instead, you can prioritize the call to action (CTA), which you might find better motivates the customer to move down the page and complete their purchase.
I think both of these breadcrumbs designs are valid, though. So, it might be worth A/B testing them if you’re unsure of which would lead to more conversions for your visitors.
7. Format the Checkout Form Wisely
Good mobile checkout form design follows a pretty strict formula, which isn’t surprising. While there are ways to bend the rules on desktop in terms of structuring the form, the number of steps per page, the inclusion of images and so on, you really don’t have that kind of flexibility on mobile.
Instead, you will need to be meticulous when building the form:
Design each field of the checkout form so that it stretches the full width of the website.
Limit the fields to only what’s essential.
Clearly label each field outside of and above it.
Use at least a 16-point-pixel font.
Format each field so that it’s large enough to tap into without zooming.
Use a recognizable mark to indicate when something is required (like an asterisk).
Always let users know when an error has been made immediately after the information has been inputted in a field.
Place the call to action at the very bottom of the form.
Because the checkout form is the most important element that moves customers through the checkout process, you can’t afford to mess around with a tried and true formula. If users can’t seamlessly get from top to bottom, if the fields are too difficult to engage with, or if the functionality of the form itself is riddled with errors, then you might as well kiss your mobile purchases (and maybe your purchases in general) goodbye.
Crutchfield shows how to create form fields that are very user-friendly on mobile:
As you can see, each field is large enough to click on (even with fat fingers). The bold outline around the currently selected field is also a nice touch. For a customer who is multitasking and or distracted by something around them, returning to the checkout form would be much easier with this type of format.
Sephora, again, handles mobile checkout the right way. In this case, I want to draw your attention to the grayed-out “Place Order” button:
The button serves as an indicator to customers that they’re not quite ready to submit their purchase information yet, which is great. Even though the form is beautifully designed — everything is well labeled, the fields are large, and the form is logically organized — mobile users could accidentally scroll too far past a field and wouldn’t know it until clicking the call-to-action button.
If you can keep users from receiving that dreaded “missing information” error, you’ll do a better job of holding onto their purchases.
8. Simplify Form Input
Digging a bit deeper into these contact forms, let’s look at how you can simplify the input of data on mobile:
Allow customers to user their browser’s autocomplete functionality to fill in forms.
Include a tabindex HTML directive to enable customers to tap an arrow up and down through the form. This keeps their thumbs within a comfortable range on the smartphone at all times, instead of constantly reaching up to tap into a new field.
Add a checkbox that automatically copies the billing address information over to the shipping fields.
Change the keyboard according to what kind of field is being typed in.
For starters, the keyboard uses tab functionality (see the up and down arrows just above the keyboard). For customers with short fingers or who are impatient and just want to type away on the keyboard, the tabs help keep their hands in one place, thus speeding up checkout.
Also, when customers tab into a numbers-only field (like for their phone number), the keyboard automatically changes, so they don’t have to switch manually. Again, this is another way to up the convenience of making a purchase on mobile.
Amazon’s mobile checkout includes a quick checkbox that streamlines customers’ submission of billing information:
As we’ve seen with mobile checkout form design, simpler is always better. Obviously, you will always need to collect certain details from customers each time (unless their account has saved that information). Nonetheless, if you can provide a quick toggle or checkbox that enables them to copy data over from one form to another, then do it.
9. Don’t Skimp on the CTA
When designing a desktop checkout, your main concerns with the CTA are things like strategic placement of the button and choosing an eye-catching color to draw attention to it.
On mobile, however, you have to think about size, too — and not just how much space it takes up on the screen. Remember the thumb zone and the various ways in which users hold their phone. Ensure that the button is wide enough so that any user can easily click on it without having to change their hand position.
So, your goal should be to design buttons that (1) sit at the bottom of the mobile checkout page and (2) stretch all the way from left to right, as is the case on Staples’ mobile website:
No matter who is making the purchase — a left-handed, a right-handed or a two-handed cradler — that button will be easy reach.
Of all the mobile checkout enhancements we’ve covered today, the CTA is the easiest one to address. Make it big, give it a distinctive color, place it at the very bottom of the mobile screen, and make it span the full width. In other words, don’t make customers work hard to take the final step in a purchase.
10. Offer an Alternate Way Out
Finally, give customers an alternate way out.
Let’s say they’re shopping on a mobile website, adding items to their cart, but something isn’t sitting right with them, and they don’t want to make the purchase. You’ve done everything you can to assure them along the way with a clean, easy and secure checkout experience, but they just aren’t confident in making a payment on their phone.
Rather than merely hoping you don’t lose the purchase entirely, give them a chance to save it for later. That way, if they really are interested in buying your product, they can revisit on desktop and pull the trigger. It’s not ideal, because you do want to keep them in place on mobile, but the option is good for customers who just can’t be saved.
As you can see on L.L. Bean’s mobile website, there is an option at checkout to “Move to Wish List”:
What’s nice about this is that L.L. Bean clearly doesn’t want browsing of the wish list or the removal of an item to be a primary action. If “Move to Wish List” were shown as a big bold CTA button, more customers might decide to take this seemingly safer alternative. As it’s designed now, it’s more of a, “Hey, we don’t want you to do anything you’re not comfortable with. This is here just in case.”
While fewer options are generally better in web design, this might be something to explore if your checkout has a high cart abandonment rate on mobile.
As more mobile visitors flock to your website, every step leading to conversion — including the checkout phase — needs to be optimized for convenience, speed and security. If your checkout is not adeptly designed to mobile users’ specific needs and expectations, you’re going to find that those conversion rates drop or shift back to desktop — and that’s not the direction you want things to go in, especially if Google is pushing us all towards a mobile-first world.