You need to do more than look at analytics data in the rearview mirror; you must act on that data to improve your site based on your overall business goals.
For many organizations, this is easier said than done. In working with companies trying to break from the past and make better use of data, I’ve found a number of common barriers that seem to arise nearly every time. Once we overcome these, progress moves at an incredible pace. Momentum builds, and the changes that are made begin to have a measurable effect on the bottom line.
Today, some common barriers to using analytics data.
Determine What to Focus on First
Companies often have a mass of analytics data from their tool, but they don’t know how to triage this information to prioritize their most fruitful site optimization opportunities.
Start by defining the site’s overall goals and the desired visitor behaviors that support those goals. Once these are identified, determine which ones could potentially affect the bottom line the most.
Is there a conversion funnel that drops a significant percentage of people at a few key steps? Is there a form with a low submission rate or a key offering visitors aren’t seeing or clicking on?
By estimating the potential financial upside of changes to these consumer touch points, you can determine which to target first. The one with the greatest potential gain should be where you start your optimization efforts.
Determine What Process Changes to Make
Now that you’ve identified the portion of the site with the greatest potential for improvement based on your overall business goals, determine why the pages may not be performing up to par. Sometimes, just reviewing the process reveals potential problem areas. Other times you may want to:
- Review where people go after they bail out of the process. This can help identify questions they may have. By understanding the questions, the page can be better tuned to address them.
- Perform a usability study or conduct an expert review of the pages or process.
- Run surveys on the site to understand what people are thinking.
Determine How to Roll Out a Change
The best way to take action on the information you’ve learned about that important process or touch-point is via testing. There are two primary ways to test your ideas for optimization:
- A/B testing. Test different versions of the same page or item against each other. You may test four different versions of the same form page against the existing page to determine which will convert the highest percentage of qualified leads. After enough traffic has passed through the five versions of the page to make the differences statistically relevant, you should be able to identify the best versions. Depending on what you find, you may want to continuing tuning and testing the pages to improve performance even further.
- Multivariable testing. This is similar to A/B testing but allows for you to test numerous small changes to a given page or process. In setting up a multivariable test, create a plan to test the different page components you want to change. At the end of the test, the ideal version can be put together based on the performance of the different components tested.
Often, some of the largest, most common roadblocks come from the IT department or the people responsible for managing the servers and rolling out new content. A/B testing is different from the way most companies usually roll out new content or changes to a site.
The first time you mention running a test of concurrent page versions, you’ll likely be told, “It’s not possible with the way the site is set up.” In every case, we can set up and run tests for clients — even if we were first told it wasn’t possible. This leads to our next topic…
Get Buy-In on Testing’s Value
It’s imperative to get buy-in from the powers that be on testing’s value. Live testing requires different rules than most companies are used to, including more frequent updates and rollouts to the site. Trying different things may not seem intuitive to these people at first.
By quantifying the test’s potential value and subsequent improved performance, and by getting management support for the process, you can begin to break down roadblocks and streamline the optimization process. This ultimately leads to a more a successful test and an organization that’s more open to testing and optimization.
You’ll run into a few roadblocks. But once you get through the first test and optimization and demonstrate the changes’ effects, subsequent tests will likely be smoother. Don’t get discouraged when you’re told it’s not possible. Keep pushing. Once you get through the barriers, testing’s rewards will far outweigh the work it took to get there.
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
In today's multichannel world how can marketers use data to ensure the experience a customer receives is relevant to them?
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more