You may already have ample site traffic, but if you aren't seeing enough conversions it's likely that there are gaps in your funnel that aren't being adequately addressed. But how can you determine where those gaps are and how to fix them? We've laid out four proven steps to achieve better conversion rate optimization.
1. Assess and Hypothesize
As with many challenges, the first step in conversion rate optimization is to get the lay of the land. Start by sifting through your analytics platforms in search of patterns (positive and negative). Both Google Analytics and HubSpot can provide you with a high-level understanding of what's happening on your site. By looking at KPIs such as bounce rate, session time, abandonment rate and behavioral flow, you can begin to track visitor movement from one page or piece of content to the next — and locate interruptions therein. More importantly, you can begin to form hypotheses to guide your experiments.
Break down where incoming visitors are starting, what pages they're clicking on and where (if anywhere) those pages are linked to. If prospects aren't following your intended conversion pathways, consider possible reasons why. Are there contextually relevant CTAs on every page that make it clear where the reader should go next? Comparing top-performing pages with pages that are seeing little to no traffic or experiencing inflated abandonment rates can give you a good sense of what's working and what's not. What major differences stand out? Does the conversion strategy and brand message remain coherent between each touchpoint? Depending on your specific challenges, this phase of your experiment will involve some detective work to help you determine which factors are having the greatest impact on your conversion performance.
2. Locate Problem Areas
As you track visitor behavior from one page to the next, trends should surface as to where prospects are defecting or deviating from your intended buyer's journey. In Google Analytics, anything over a 70 percent bounce rate should indicate that there's room for improvement. Heat map tools such as Hotjar can help inform your hypothesis about visitor behavior. Are prospects attempting to click on an item that isn't an active link? If so, a heat map will indicate heavy activity around that area. What's more, heat maps can give you insight into what visitors intuitively want from their web experiences. In short, if you find that prospects are trying to navigate by clicking a button that doesn't exist, why not make a button and place it accordingly?
As with any experiment, altering too many variables at one time will make it difficult to determine what's having the greatest effect on your conversion rate. Try to limit your changes to tracking one or two factors at a time (such as adding relevant, in-text CTAs to a page that's not performing up to expectations) and measure the effect after a defined trial period.
With A/B testing, it's important not to get too caught up in the minutiae of manipulating variables. For example, though the color of a CTA button might have a slight affect on click rate, it's not likely that altering a single color will be the answer to a low CRO, or that it will inspire a change significant enough to compensate for more fundamental gaps in your funnel. In that same vein, if you find your issue is design specific, it's likely you'll eventually reach a plateau — a point where no degree of manipulation can get you to where you'd like to be. Like replacing faulty parts in an old car, there should eventually come a point when you weigh the effectiveness of your efforts and investment against the potential return. If the cost of fixing problems is beginning to eclipse the cost (and projected ROI) of investing in a new model, then you're likely headed down a dead-end road. In such instances, you should adjust your course — which may mean considering changes like a complete site redesign.
All data has an expiration date — and even your highest performing pages will not remain so without attentive care and the right contextual modifications. In order to continue to grow, your CRO testing process needs to be just that: a perpetually repeating process. Even if your experiments inspire the opposite result of what you would like to see, take each test as a learning experience rather than an absolute win or loss. Aside from changing web variables, think outside the box by employing other cadences or strategies and measuring their effect on stick rates and conversion wins. Things like email drip campaigns can peak interest in PCOs and promote more sustained engagement with related web content, for instance.
Additionally, offering greater customization can never hurt. When in doubt, return to your trusty persona-based segmentation strategies and tweak content to better align with your target audience's unique challenges, goals and desired solutions. Furthermore, make sure that all your communication — from email content to landing pages, PCOs, homepages and social media copy — is congruent with your tone/persona goals and is written with a specific conversion strategy in mind.
Meryl is a former New Breeder.
Connect with the author