One of the questions we hear from our testing clients most often is, “Why aren’t you measuring the lift provided to conversion rate?” It’s a good question! When my kids were little, one of my favorite stories to tell them was called “The Chain Reaction”(a photocopied page with laminated visuals–you won’t find it published). It told the story of how Mr. Bunny forgot to kiss Mrs. Bunny goodbye when he left for work which upset her, so then she snapped at Tommy. Now, because Tommy was upset, he pulled his sister’s hair, who then griped at her friend on her way to school, and so forth. Everyone had a bad day because one person was thoughtless. It goes on to articulate how one person turned it around, said sorry and the chain reaction went back the other way, making everyone’s day better again. I loved this story because it was supposed to teach my children the effect their actions have on others for good or bad. Sometimes we are tempted to think about conversion optimization this same way. If I improve one thing, it will turn everything around and increase my eCommerce conversion rate. We like to improve things , and while we hope this is true, conversion optimization is more like baking cookies. If you are given a cookie recipe you know is great, many people will still make changes they hope will make it better.
If I give you a bad recipe, the cookies will be bad. I can fix some things in the recipe, but who knows if I fixed everything? Having the right ingredients in the right amounts does not protect you from over baking or over mixing the cookies. If we can take measurements along the way, we can more accurately gauge at which step the process might be going wrong. Taste the cookie dough. Measure the oven temperature. Check the recipe against known recipes that work well. The more places we can check the process, the better our information.
Checking the progress at the point of change is the standard, but like all good rules, there are some exceptions. Occasionally, there is a change that has reach beyond the simple micro-conversion for that page. In cases like these, it makes sense to compare overall ecommerce conversion rates. We recently came across an example of this; the question was, “where in the checkout process should we show the shipping information to our client?” Hiding it early on could make those pages progress better, but then do customers jump ship when it is eventually shown? Or, maybe users are frustrated at not having the information up front and hiding it makes the page convert lower. We answered this question by deferring to overall ecommerce conversion rate from the cart page we were testing instead of the micro-conversion from cart to checkout.
Conversion Optimization is all about measuring and tracking these changes so we know whether it was an improvement. In conversion optimization, we call these checks add to cart rates, progression rates, micro-conversion rates, and best practices. We might also evaluate overall bounce rate and ecommerce conversion rate, but these are much harder to pin down. Although it’s tempting to ask how each change made affects ecommerce conversion rate, in all but a few circumstances, we really need to ask ourselves about the metric attached to the page we changed. If we make a change to a product page we SHOULD ask ourselves, “how does this affect the add to cart rate?” Not, “How does this affect my ecommerce conversion rate?” Making a change without testing introduces further variables. Testing allows us to compare the add to cart rates of the old page versus the new page at the same time in the same ecosystem. This eliminates the need to account for variables introduced before and after the current page and evaluates the page solely on it’s primary goal.
Each page on your site should have a goal. Progression or Micro-conversion rate refers to the rate at which this page accomplishes the stated goal. Improving each page to more effectively reach the stated goal, brings you closer to having all the pieces in place to bake a great cookie-or website!