Find CRO low-hanging fruits
1. Identify UX improvement opportunities starting at the bottom of your sales funnel from product to checkout pages for ecommerce, and forms in lead generation.
Prioritize the opportunities by ease of fix and closeness to the conversion. You can get even more specific. For ecommerce, improve upselling and cross-selling on the product-page-to-cart transition, which can help bump average order value.
2. Optimize your social proofs and test them in different locations on the page.
Most social proof elements need minimal effort, or slight design intervention. For example, adding photos to existing testimonials from clients or user reviews. What may take more time, is to customize social proof elements for each channel or audience.
3. Tighten up and test your value proposition copy, headings, and placement.
The low-hanging fruit of copy changes in testing different value propositions, headlines, and offers. Those choices may seem obvious, but too many companies are oblivious to their shortcomings. Instead of highlighting product features, what the product does, use your copy to talk about what the product does for your customers, the benefits they’ll get from buying. Position your benefit-oriented value proposition copy where users can easily see it. Don’t bury it under other information. Use the homepage hero area to display one clear offer instead of several competing offers, carousels or other distracting things that may make it more difficult to understand what the site is actually about. For example, Trello uses its headline to highlight its benefit-oriented value proposition. The features that deliver those benefits are below, in the subhead:
4. Change your call to action button to a high-contrast color and check that it's easy to find.
Don’t waste too much time on button colors, pick something that’s high contrast and move on. Position your CTA where more people will be able to see it. For instance, if your CTA is buried under 10 paragraphs of text where fewer users are likely to scroll, you want to move it to a more prominent position. Test if the new position, somewhere up the page, will increase conversions.
5. Add microcopy that explains why and how to fill out a form, and make calls to action descriptive.
For ecommerce, remove unnecessary fields on your form and add microcopy to improve the clarity of the fields format, like phone number and zip code. For lead gen sites, set forms into a sequence with small, simple questions to start, which you can use for personalization later on. Optimize the microcopy for all your CTAs, so it tells people exactly what happens when they click a button. For example, a Reserve Now button on an accommodation booking site, should lead you to a reservations page, not a listings page.
6. Run user tests, session recordings, polls, or surveys about your site, then cross-reference all your data to find patterns and broken website elements.
If you have the budget, run user testing videos that are about 30 minutes long. Do a round of 10 videos, then ideally repeat the process after a while to obtain insights you can compare against. If you’re working on a lower budget but want to stick as closely as possible to user testing, do another round of research but replace user tests with session recordings. This can help you identify some of the same technical or UX issues and assess where users pay attention. Use a tool like Hotjar for running session recordings, heatmaps, or on-site surveys. Alternatively, use surveys and polls to obtain insights about ways to improve UX, increase motivation, or lower friction. Include at least 8 questions, and ideally, aim to collect at least 200 answers per question. If you have a decent email list, survey your customers also via email. Try to find out about their pains when they use your site, to come up with new test ideas or create testimonials or tweak site copy. Multiple data sources are key to spotting patterns. If the same issue comes up in a user test and on a survey and in analytics, you can feel more confident about a test hypothesis. You’ll also have a stronger case to make for any approvers. Take advantage of analytics to get more clarity on KPIs, especially on lead quality and LTV. Use cohort analysis where applicable.
7. Fix pages and elements that don't meet user expectations or are broken, without going through a formal hypothesis testing process.
An example of elements not meeting user expectations: a company’s product pages had rating scores across all products that looked like clickable reviews, similar to Trustpilot ratings. Heatmaps showed that visitors kept clicking on these ratings and survey insights showed the need for them, but there were actually no reviews or testimonials to read on the site. Proper site speed is another user expectation. Test site speed yourself. If it takes seconds to load, it’s an issue.
8. Avoid major redesigns, wild ideas, and rushed decisions to change vital site elements.
Whether looking for quick wins, big wins, slow wins, small wins, or anything else, major redesigns are expensive, time-consuming, and flat-out risky. Within the framework of iterative redesigns, some changes are a heavier lift or have uncertain value: Combining multiple hypotheses into one test. Product-customization tools, user dashboards, and other things essential to the user journey that require a lot of backend work. Testing a completely new feature. Avoid blindly copy-pasting quick-fix ideas that worked on one page, and which will not necessarily work on another page. Be aware that the low-hanging fruit might not be in the first places you’d look, because other people might have already fixed those items.