You probably also recognize the need to test your assumptions with experiments and split tests, that you need to overcome objections, and, well, convince them to convert. Sure, that eventually turns into tests, data, and a road forward. And it means that most of us are, at best, making decisions based on a half-finished model.
Sure, over half of US consumers with net access are regularly shopping online. But most sites have conversion rates under 5 percent , and unoptimized sites are lucky to capture value from more than 2 or 3 percent of their visitors. Compare that with physical retailers, and it starts to look absolutely horrendous. That last point is a big problem. Some of them talk about awareness, interest, desire, then action, or use different terminology.
But they all take the same basic framework: Do these steps leave you feeling uninformed? Unsure how you could possibly use them to your benefit? The steps just sit there. But how do we advance them through those stages, instead of sending them running for the back button? After analyzing academic papers on CRO from to , the researchers concluded that the buying funnel model was missing one thing: First off, in the three case studies, applying just 5 of these web experiences was enough to double conversion rates 2 out of 3 times and make some serious cash in the third.
But the breakthrough comes from combining these two ways of looking at user behavior into one unified framework. At each stage in the buying funnel, the user is experiencing one or more of these elements. This helps us pinpoint which kinds of changes are worth testing during each phase in the buying funnel.
Now, at each step in the buying funnel, we know which elements are most important to test. We can ignore the catalyst when the user is making the actual purchase, and afterward. By combining these models, we have a framework for deciding what to test, why, and when. When you understand what motivates the user to visit the page, you can design accordingly. This is why user targeting is so important. Take a look at:. This is what people tend to think of when they talk CRO: An understanding of visual psychology and design principles can pay dividends here.
This one is huge. To establish value, you need to accomplish three things:. Keep in mind that usability testing is a very different beast from split testing. Once you fix those obstacles, you iterate. Each iteration has a small sample size, but over time this sample size should get fairly large. Unbounce has a post on the complicated relationship between usability testers and split testers , and Smashing Magazine also has a helpful guide to improve your usability testing.
This one gets missed a lot. You can adjust button colors and directional cues all year long and get nowhere if your core approach to persuasion is flawed. Persuasion is based on two fundamentals:. When you compare online and offline shopping, there is a serious confidence gap.
This is one of the absolute biggest gaps between the two worlds of retail. Online stores are both more and less interactive than physical stores. When you physically drive to a store, you can directly ask questions and get help without an awkward interface.
So, how can we approach this actionably? It actually all boils down to a fairly simple process:. It gives you principles to base your decisions off of, instead of just throwing things at the wall to see what sticks. We need to address what is going through their minds during each stage, and why. Once we recognize which level of experience matters and when, we can eliminate objections before they happen, keep users on the site, and even keep them coming back for more.
Have you ever addressed the wrong thing at the wrong time? Does this model clarify what went wrong? Your criteria is an aid in our daily work and is very accurate, since it explains perfectly how consumers behave and how to adapt the contents of the web around that behavior. Congratulations on your post! However, I do think your title is a bit misleading.
Great post, Pratik and well researched. It was nice to include the case studies as well in order to prove the theory. Hi Pratik, Awesome post!
Though title is a bit misleading! Log In Start Free Trial en de es br. Most of us have no idea what our users are thinking about.
Hit him up on Twitter for a quick chat. Great write up, Pratik. Working to implement into some testing now! I enjoyed this post, and like this framework presented. More has to be said about this, and I enjoyed the post. Thank you for a good post, to the point and with hands-on advice.More...