Landing pages are a great place to focus optimization efforts, but where do we start? A successful testing program depends on a data-driven approach. Here is an approach using examples from Google Analytics and Google Optimize to identify opportunities, understand your audience, analyze user behavior and design impactful landing page tests.
1) Identify Opportunities
Start with analytics to identify landing pages that offer opportunities for optimization.
How do I identify landing pages that might have optimization potential?
- Traffic. Volume is required for a test to yield a statistically significant result and have a measurable impact on your bottom line.
- Conversion rates. Pages with lower conversion rates may present your most meaningful opportunities. If they also have a decent volume of traffic, include them in your list, though some may get ruled out in the next steps.
- Purpose. Some landing pages with high traffic and low conversion serve a purpose other than conversion and should be excluded from your short list. For those that remain, the next steps will help you assess their potential even further and begin the ideation process for your first tests.
2) Understand Your Audience
Once you have identified some landing pages with potential, it is important to understand the audience of these pages. Often, it is the nature of the audience that contributes the most to a poor conversion rate. Understanding and segmenting the audience is critical to confirming which pages present the most opportunity, as well as ideating about how best to optimize.
How can I understand the audience of my highest-opportunity landing pages?
- Convertible? Assessing your audience for convertibility starts with ruling out segments that are not likely to convert, such as existing members or users visiting the site to leave a review. Relook at traffic and conversions for your landing pages with these segments excluded.
- Relevant? Assess relevance of traffic sources for hints about user intent. Keywords can be assessed for some channels to determine whether irrelevant searches are driving less convertible traffic. Referral sources can be reviewed to form opinions on the relevance of traffic generated.
- Controllable? Depending on the source of the traffic we may not have much control over which pages users land on. Here are some examples of how this information can be used in your assessment:
- If there is a high volume of irrelevant traffic driven by organic search, which is not very controllable, you might choose to exclude that traffic using a Segment in your landing pages report. Excluding this traffic at this stage, and later while analyzing test performance, will give you a clearer view of the conversion rate of your relevant audience. You might find that some of your landing pages are performing quite well in this light and remove them from your opportunity list.
- Where the audience is controllable, for example paid search, your options and next steps will differ.
3) Form a Data-driven Hypothesis about What is Wrong
With a refined list of opportunity landing pages, and defined segments to narrow in on your convertible audience, let user behavior guide your optimization effort.
What should I inquire of my data? How can I leverage Google Analytics to answer these questions?
- Am I serving the most appropriate page/content for my audience? After your audience assessment you should have a pretty good idea of whether the page is the most relevant for your audience.
- Are users starting the conversion funnel? It can be helpful to configure a Google Analytics Goal for conversion funnel starts. This will allow you to quickly assess ‘start rate’ from a landing pages report. A strong start rate might lead you to assess the funnel itself rather than focusing on the landing page. A weak start rate would drive you to further assess the page.
- Are users bouncing and if so why? To better understand a high bounce rate for a relevant convertible audience look for contributors such as:
- Obvious problems on the page (404, broken images, etc)
- Slow load time (Google Analytics Page Timings report)
- Scroll depth. Not available out of the box in Google Analytics but there are several handy ways to implement it. Get help with scroll depth tracking
- CTAs (Calls to Action). Are the next steps clear and functional?
- Technology factors. Are certain devices or browsers contributing to bounce rate disproportionately? If so, use Segments to isolate the poor performing technologies and recheck the above steps.
- Tracking issues. It is worth noting here that if a user’s next steps are not tracked appropriately in the same GA property, bounce rate can be artificially inflated. Get help with tracking issues
- Are users navigating elsewhere to find what they need? The Navigation Summary report will show you where users are navigating to. Evaluating their next pages should reveal hints about what they are looking for and not finding on your landing page.
At this stage, you should have a very good idea about where to focus your efforts. You may have even identified some issues that can be fixed quickly before embarking on your testing journey.
4) Design Test
Test design should be driven by the observations you have made, then targeted and measured against the relevant audience segments you have defined.
Here are just a few examples of how to translate your understanding of user behavior into sound test ideas.
- Relevance. When users are not seeing the most relevant page or content you might focus optimization on:
- Redirecting to a more appropriate page.
- Personalizing content based on what you know about the user.
- High Bounce and High Load time. Focus optimization on page speed.
- High Bounce and High Scroll Depth. Focus optimization on presentation of your call to action.
- High Bounce and Low Scroll Depth. Focus optimization on page design and/or content.
- Strong Conversion Path Starts. Focus optimization on the funnel.
High Navigation. Focus optimization on clearer communication. Or identify strong performing navigation and make it more convenient.
Analytics and Testing go Hand in Hand
Analysis is the key to driving an efficient and effective testing program, this is true regardless of the tools used. A reliable integration between your analytics and testing tool can offer even more. For example, using Google Analytics and Optimize together you can:
- Save implementation time by using goals you have defined in Google Analytics to measure your tests in Optimize and benefit from a single source of truth.
- See test experience data in your Google Analytics reports via built-in custom dimensions.
- Target and personalize using audiences you have defined in Google Analytics with behavioral data collected there. (Google 360 only)
Analytics and testing are essential to improve your business landing page conversion rate. At Empirical Path, we can help your business audit web analytics implementations to ensure crucial visitor behaviors are captured accurately and reported clearly. Let us help you create a strong testing program saving you time and money.
Tidy your data, transform your business.
how we can help you.