Testing and Optimization is a Full-Time Job
March 29, 2016
Testing and Optimization is a Full-Time Job
Designing and maintaining a website that delivers great results and conversion rates is not a one-time job, but an ongoing journey.
As your website keeps growing with additional landing pages, web forms, home page updates, product and service pages, shopping carts, checkout pages, search pages, and more—it becomes imperative to look at each of your pages with critical eye, identify areas that can make a difference, and conduct tests to identify the best options for success.
But before we dive into the “how,” let’s first clear the air about what optimization and testing are, and why some people might have a misunderstanding or three about the concepts.
Let’s start off with a HUGE statement:
Web optimization is not something you do only once.
Or five times. Or six times a year. Instead, it is an ongoing process of optimizing the key pages on your website, because as long as things change in your business, your site needs to reflect these changes. (Just ask our friends at Optimizely.)
Have you added new products or services? Change your target audience? Tweaked the logo ever-so-slightly? Then your web presence needs to reflect these changes in value and offerings…on every last page of the site.
Keeping a site optimized facilitates an engaging experience by keeping it on-brand, on-message, and current with your product offerings. You certainly wouldn’t allow your LinkedIn company page to have outdated information, right? Your web audience won’t allow for it, either.
An optimized site presents a unified brand identity that is cohesive with social media, email outreach, and print collateral. This will build trust with visitors, and will help keep bounce rates under control, improve the depth of each site visit, increase time on site, and bring visitors closer to conversion.
But, in order to achieve these things, you first need to do the legwork and figure out what your customers want and need, and what you need to do to deliver it. (Hint: start here).
Optimization for any site—regardless of purpose—should involve:
Knowing what’s already working. Change for change’s sake can be just as damaging as not making changes at all. If your homepage shows consistent click and conversion rates, then it might not need much further optimization. This is not to say you shouldn’t test. (In other words, ALWAYS TEST). But simply moving things around or changing words won’t always deliver the results you want.
Clearly defining objectives. Before you begin optimizing your site, it is important to understand the objective of each page and how well these objectives sync with your overall sales goals. Also, be mindful of visitors arriving via different marketing channels.
For example, when you know your macro goal is to have a prospect fill out your contact form, a landing page designed to simply capture leads likely doesn’t need three scrolls before hitting the conversion form. State your case, explain the value of the opt-in, and make it unbelievably easy to complete the task.
Reinforcing value proposition. Again, any pages that don’t adequately reflect your company’s core values need optimization. From minor shifts in design and font choice, to unwieldy URLs, and slight alterations in page flow—all of these factors can kill conversion.
Providing a cohesive experience. The bottom line is, if your readers get even the SLIGHTEST feeling they’ve been removed from your core website experience—i.e. seeing an e-commerce platform name in the address bar instead of your URL—their first instinct will be to back out of whatever conversion they were about to make.
Restructuring the content and calls-to-action. Chances are, the first culprit in your conversion and retention problems stems from content and CTA placement. If your content is too long (or too short, for that matter), your CTAs are hard to find, your image use distracting, or any combination of the above, look to rectify the situation.
Again, keep the writing to the point, and the page flow as scannable as possible, so readers can find the info they need, or the value they require, to move ahead toward conversion.
Test. Test again. Then test forevermore.
We’re not breaking any ground with this statement, but:
Testing really is the key to any and all optimizations.
Just trying new things without parameters for your experimentation, “just to see what sticks” isn’t only lazy, but could also cost you potential sales.
While you’re tinkering with images on third-level pages, your competitors are testing their sites through numerous variables to achieve their goals.
Make no mistake, sometimes results can come from tinkering with low-level pages. But this can’t be truly determined without testing the page elements in question.
A|B (split) testing is a continuous process of running simultaneous experiments between two or more page layouts or content elements (headline, call-to-action, banner, etc.) to evaluate which performs or converts the best. When the test runs, the testing solution should randomly assign site traffic to each variant and give equal weight to both variants, i.e. 50:50.
A|B testing seems simple, because generally, it is. But when you find that one key tweak, change, or stumbling point, it’s as good as gold.
In fact, when done right, A|B testing can boost conversions by 25%.
It’s important to note that a “one and done” approach rarely yields quality results. Instead, there are tools and resources to help you determine the appropriate A|B testing length and duration for statistical significance, or, point you in the direction of multivariate testing to achieve more focused results.
(Keep in mind that test “length” and “duration” are not always the same thing. In some cases, the test’s length refers to how long one specific A|B matchup lasts, while duration refers to how many times you might test an “A” vs. a series of “B”s.)
Web tools (we love using this Sample Size Calculator from Optimizely) are available to test your ideas and help you learn what gets your audience to take the desired next step, alongside what discourages them from taking action. Changing just one element can show an increase or decrease in the performance of a page.
Multivariate testing, in comparison, allows you to test many changes simultaneously. You’ll get accurate results without having to increase your total sample size, and you will be able to identify the impact of each individual change that you have made to optimize your page.
There are many advantages of multivariate testing. As the term implies, many variations can be tested at the same time. While determining these variables can be imposing, once testing is underway, modifying different elements in unison can help produce fast, definitive results.
For web and app optimizations, where countless combinations of elements are needed to create a cohesive experience, multivariate testing allows marketers to find that unique “special sauce” that converts prospects into leads, in ways that split tests couldn’t do in the same timeframe.
And it’s not just the interactive pieces that matter. Multivariate testing also makes it easier to test different versions of elements like headlines, content placement, ad placement, images, and more.
Key testing processes
So, now you know what types of tests to run, but aren’t sure how to get started. Well, as you likely know, testing is an ongoing process, not something to be done in a silo. While the “old guard” of marketing will regale you with stories of winning marketing campaigns built solely on instinct and gut feeling, these tales never seem to mention the millions of dollars wasted on campaigns that aren’t properly researched.
Before you begin a test or series of tests, make sure your plan is properly vetted and…well…tested, before you sink countless hours and dollars into a doomed effort.
- Discovery: Start the optimization process with a discovery session with a strategist, content manager, digital marketing manager, and e-commerce manager to discuss site objectives, value propositions, and how each page contributes to conversion.
- Behavioral audit: After the initial discovery session is over, study quantitative data in a web analytics tool. The idea here is to study user behaviors to figure out the importance of each page in the overall conversion process, and identify opportunities for conducting split or multivariate tests.
- User test: Get real-time, in-person feedback to see what is and isn’t working based on your site objectives and goals. For example, the world’s best keywords, product descriptions, and page colors are great, but ultimately meaningless if people can’t find your “Buy Now” button. Simple user tests quickly and effectively pinpoint stumbling points to optimize.
- Selection: Once the importance and potential of each page is established, evaluate selected pages for testing. Formulate an A|B and multivariate testing strategy for the problematic page(s) on your website and share it with your team to begin executing. (Hint: this is where Optimizely can help you get started.)
Conversion rate optimization (CRO)
Much like overall site optimization, CRO is a continuous process of improving the key sections, pages, and flow(s) on your website, only laser-focused on pages and elements specific to conversion. This drives visitors closer to the conversion path by eliminating unnecessary steps or roadblocks in the buyer journey that might deviate or distract them.
It is necessary to do a full website audit, from a usability perspective, for online businesses to kick off a CRO initiative. Additionally, it’s important to evaluate the various sources of traffic to determine how to expedite conversion. Then, after an assessment of conversion goals, conduct the following steps:
- Site architecture audit: Focus on the information architecture, site taxonomy, and the various conversion funnels on the website. Competitive analysis could be helpful here.
- Conversion audit: View your site’s web analytics tool, heat maps, and online survey tools. Also look at the performance of ongoing A|B or multivariate campaigns on your website.
- Audit report: Lastly, use your findings to create a testing and optimization strategy that focuses on optimizing pages and funnels on your website.
Seems like a lot to remember? Never fear. You don’t have to do it alone. There are tools to help move this process along, and they’re WELL worth checking out, especially considering:
The average ROI when using a conversion optimization tool is 223.7%.
Making the most of your tests
When it comes to testing, defining success involves knowing the answer to the ultimate purpose of your site and turning it into more specific or quantifiable success metrics. This is fundamental, and arguably the most important element of any test process.
Once testing is complete, it’s important to identify any leaky stages in your goal funnels. These are the places where visitors drop off, or the areas where you might be better able to move visitors through a set of desired actions.
Also, use test results to create behavioral hypotheses that can serve as the basis for—you guessed it—more testing. Like we said at the outset of this piece, testing is an ongoing process. But it doesn’t have to be a laborious one. With every test result comes a finding that can help fine-tune further analysis. Heck, even a “non-result” is still a result of sorts, and still provides you with data to refine your future testing.
In short, once a testing culture is ingrained in your marketing teams, it ultimately gets easier as you work toward making more granular, specific adjustments to improve user experience. And just when you get it as close as possible to perfection…it’s time for a new site design to encompass emerging technologies, better offerings, and ever-evolving value statements to meet the needs of a changing audience base.
Testing and optimization doesn’t have to be complex. Success in these areas hinges on the delicate and complex task of finding out what customers want, and when and how they want it, as their needs change constantly during their journey through the sales funnel.
Applying your learnings to all your marketing channels actively and effectively tells you what’s working and what isn’t, so your sales and marketing teams can focus on the efforts that have the biggest impact on your bottom line.
To learn how Lucid Fusion can help you optimize your web and mobile assets (and maximize your ROI along the way) contact us today!
Use the download button below to download a free copy of this article.
Download your free copy of this white paper