
This is potentially an unpopular opinion, but we're okay with that. What we've found over the years is that metrics are incredibly helpful, but when devoid of context, they can actually be misleading.
Here's just one example.
One of our clients had 8 separate eCommerce funnels for their product. At the end of each checkout path, users were required to watch a 3-minute video. There was no way around the video. No skip. No "continue" button. Just a video that you were destined to watch.
When I first saw this checkout flow, I was aghast.
Are we really forcing people to watch a video, even after they've committed to purchasing?
Here's where the numbers lied
Now, you would think that many people would abandon their cart right here. Who has time to watch a 3-minute required video? But they didn't.
So the executives thought the video was fine. They thought: It's not holding anyone up. If anything, it advertises more products to customers, and could lead to more upsells.
But here's the kicker.
People were watching the video, not because they wanted to. But because they were genuinely unsure if their credit card was ever charged. You see, the checkout process was so convoluted that people didn't know if they had already bought something or not. And then the video popped up.
Yes, the video was designed to upsell additional products. But no, people didn't actually want to see it. In fact, the Hotjar click tests revealed that people were trying to get out of it, or find a way around it. But they stuck around because they never got confirmation of their purchase.
So the entire customer experience was both confusing and annoying, but from the numbers alone, you'd never know it.
Proceed with caution
This is where context is king. It wasn't until we went through the checkout process ourselves, analyzed the numbers, walked the customer journey, and felt the customer pain - that we actually discovered the real problem.
The numbers made it look like things were fine. Understanding the user journey showed us that they weren't.
So yes, dive deep into the data. We love numbers and metrics here, too. But if that's all you're staring at, you might be missing the forest for the trees.
Note: Not a single word of our Insight articles are written with AI. We're keepin' it real, folks.
Want to work with us?
Let's talk