Building and Testing an Optimization Plan

Do you want to hear a secret? You can guarantee more conversions! How? The key is following a well-defined, systematic 6-step process. Just keep reading…

When you decide that you want to start optimizing the conversion rate of your website, you have two choices: apply popular examples of split testing experiments that you found online, or build and test an optimization plan that fits your strategic needs.

By reusing split testing ideas in your site you try to mimic the expected results of those experiments in order to obtain the same gains. However, you do that without noticing that it is not possible to generalize the results to all situations. The described results of a popular split testing idea just hold for the conditions in which they occurred. The particular page design, the concrete site where the experiment was run, and the audience that viewed the variants and (unconsciously) decided which one performed better than the other. Maybe you are lucky and the changes you make work well in your site. Maybe not.

On the other hand, by building and testing an optimization plan you follow some intelligent steps that make you think about your business needs, the way your website is aligned with them, and what’s currently happening in there that’s preventing to increase your conversion numbers. In some way, you follow the same approach scientists do when do research: building and testing an optimization plan is like applying the scientific method.

The Oxford English Dictionary defines the scientific method as “a method or procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses.” That’s the strategic approach you need to follow.

Therefore, to optimize your site you’ll observe, take measures, formulate hypotheses, and test them. It may seem like a lot of work, I know that. But you need to understand this is a consistent, structured, and ongoing process of making your website better over time. There are no silver bullets. Consistency is the key to success!

Let’s review the steps to build and test an optimization plan for your website.

Step 1: Understand your Business

Before starting to optimize your site, you need to identify what “conversion” means to you. You must know what you’re measuring and attempting to optimize. Are you selling products to an specific market segment? Are you a content writer and for you conversion means visitors reading your stuff? Do you want more subscribers and leads? All these questions should be answered before moving forward. There are times when some goals may be less obvious than others. Identifying them is your priority task now.

Next think about you and your competitors. What makes your company different from those competitors who sell the same or similar products? You want to convince your audience to choose you! In order to do this, you have to understand what type of audience you’re targeting, and part of this also includes understanding who you’re not targeting. Do this by creating customer personas and sharing them with everyone involved in your marketing efforts. What problems are they trying to solve? What would make their life better? Find the way that you make your target audience happy all while sticking to who you are as a company.

Step 2: Understand your Web

Here, we are focusing specifically on the website itself. You need to discover whether your web is directly aligned with your business or not. Before optimizing your website, you need to define what the goals of your site are. Pretty obvious, I agree, but you do need to be more specific than “to make money”. What your customers want to do is not always the same thing as what you want them to do. Your job then is to give them what they want and still get what you want. You must define the specific actions that you want them to take on your site. Everything on the site must have their own micro goals that contribute towards your main business objectives. You need to agree what the key goals of the website are and set up core tasks on the website to drive these goals.

It is important for you to clearly understand that before you can begin any optimization campaign, you need to understand how your site is currently performing. Use data and tools to understand where visitors go, what they look at, what makes them leave. It is also at this point that you should review funnels in analytics. Find the problems with your website. Try to find where the most common drop-off points are and take a deeper dive into why. What are the biggest objections that customers have? What is stopping them from buying? There’s a number of ways to get answers to these questions. A well planned and executed strategy can help explain this.

To effectively measure your website you need to combine three core elements; web analytics, surveys, and user testing. Analytics gives you direct facts on what is happening on-site, and most of the analytics tools are free, so no more excuses. In addition to it, customer surveys will tell you what areas need to change and (if you’re lucky) how hey should change. And directly working with users and interacting with them –that is, testing the usability of your site– will show you how they are experiencing and using the site. Heatmaps can be extremely helpful on teaching you the way visitors interact with your site. Maybe your call-to-action buttons are placed down enough in your page so that your visitors don’t see them because they don’t scroll down. That’s something stupid a heatmap can show you after tracking a bunch of your visitors. But maybe a stupid thing like that is the hole in your pocket that’s preventing your from earning more money with your site.

Nelio A/B Testing

I was very impressed by the quality of this plugin, how easy it was to set up, and the outstanding support Nelio provides. I highly recommend Nelio A/B Testing.

Josette Millar

WordPress.org

View Plans

Step 3: Form a Hypothesis

This is an important step in the testing process. It requires articulating the problem and what you believe the solution to be. Based on all of the information you gathered on previous steps, what would you like to test? Without time to interpret the information the data you have is meaningless. Find the problem areas, investigate the detail, get user perspective, and come up with potential tests. Brainstorm on ideas and look at your data! From here decide which hypotheses you would like to proof.

Do you want to reduce bounce rate? Do you need to improve the copy in your landing page? Do you need to improve the checkout experience in your e-commerce site? Stick to your problems and find hypotheses that could improve your site. But, what does a hypothesis mean?

A hypothesis is an educated guess about how things work. Most of the time a hypothesis is written like this: “If [I do this], then [this] will happen.” Fill in the placeholders with the appropriate information from your own website and business goals. As an example, if your call-to-action buttons are placed at the bottom of your page, it may be the case that some of your visitors don’t see them because they don’t scroll down your site. So a hypothesis could be: If I place the call to actions both at the top and at the bottom of my site, then at least a 25% increase in conversions will happen.

Note that your hypothesis should be something that you can actually test, what’s called a testable hypothesis. In the previous example, you expect to get 25% more conversions. That’s something you can easily count. In other words, you need to be able to measure both “what you do” and “what will happen” in your hypothesis. It is important to know what success looks like and therefore, how to tell if the hypothesis was a positive one. You need to document your expected outcome in order to check afterwards if the test was (or wasn’t) successful proving (or refusing) the hypothesis.

Step 4: Define your Test

This step is all about taking the analysis, investigation, and hypothesis that result from step 3, and implementing the test strategy that you’ll run in step 5. The process of conversion rate optimization trough split testing involves serving up different page variants in order to measure objectively which version will be more successful. While you can test smaller site features (for example, your site’s body text font), begin by testing the variables that are likely to have the biggest possible effect on your overall conversion rate.

Are the changes directly testing my hypothesis? This is the question you need to focus on when designing the experiment. Probably the easiest thing you should do is working with only one change at a time. Look for something that won’t be too complicated to change and measure, but with real potential for improving conversion rates. The more complex your variants, the more work you may need to get the test ready. If you insist on changing multiple things at a time you may never fully understand what the solution to the problem truly was!

Following the previous example, you could design an experiment where you test your original landing page against an alternative version where the call-to-action buttons are placed twice in the page (you know, at the top and the bottom). That’s super easy to create, you’ll only need to duplicate your actual page and copy the buttons at the top. So simple your mother can do it! It’s a matter of minutes, or even seconds if you use a page builder like OptimizePress or Visual Composer.

But be careful when designing the alternative versions that will be under test. Don’t be afraid to ask for a second opinion. If you designed the page, you may not readily spot the problem. You might have to lower the “risk” of designing a bad experiment.

Step 5: Run your Test

After gathering all of this information, you’ll need to load it into a tool that will automatically serve up either your original version or your alternative one at random to visitors and record the number of conversions that stem from each variant. Don’t worry about that, Nelio A/B Testing comes to the rescue. Upon completing a very easy steps, you’ll be able to launch your split test and start generating data immediately. Take a look at what you can test with Nelio.

Step 6: Learn from your Results

Now it’s time to see if our work has paid off. But wait! Before diving in and assessing if your hypothesis was correct… Was statistical significance reached?

Statistical significance is achieved when there are enough conversions, conversion rate is high enough, and the conversion rate delta is great enough. Fortunately, today’s testing tools (like Nelio A/B Testing) calculate statistical significance automatically, which simplifies the conversion champion’s decision-making (and saves hours of manual calculation!). Remember not to stop your experiment before achieving statistical significant results. Otherwise you may choose a false winner that could harm your conversions!

In addition to it, if you’ve run a test with more than 30,000-50,000 visitors and one variant is still not statistically significant over another, then your test may not ever yield a clear winner and you should revise your test plan or reduce the number of variants being tested.

Now, let’s go to the point… Was the hypothesis correct? Yes? Hooray! Substitute your original page with the winner alternative and buy a round of drinks to celebrate your success! But when the hangover wears off you’ll realize that you’re not done. Unless your site has a 100% conversion rate (is this possible?), you still have plenty of people leaving your site without converting. Then, as soon as you’ve ended one test, start another right away. There are hundreds of thousands of different combinations that can be tested on any given website. Don’t miss out on the one that could make all the difference in your business by running one test and then giving up on the power of conversion rate optimization.

But, what if the hypothesis was incorrect? The important thing here is to not give up and make sure you’ve learned something from the process. You decided to run a test because you identified a problem. Just because the first test didn’t work, it doesn’t mean you have to stop testing that area; at this point you loop back to step 4 and start again. At least you’re sure now that you didn’t introduce changes in your site without testing them, which could have led you to worst conversion rates.

Conclusion

Truly effective website performance can only come from the measurement of metrics and experimentation to determine which conditions and variables lead to the strongest conversion rates. By continually measuring and testing using the conversion rate optimization (CRO) plan described below, you can boost on-site engagement and increase site profits significantly with just a few tweaks.

Featured image by Kenny Louie