What's the biggest problem you face when A/B testing your funnel?
A/B testing is supposed to systematically isolate winning and losing acquisition funnel content.
My biggest struggle, by FAR, is understanding why people bounce from my landing page after clicking through my ad.
If they don't bounce, and enter the funnel, it's much easier to test why they don't make it to purchase. But isolating the reason for a high bounce rate is tough.
Have you faced this problem, and how have you managed it?
Replies
Can your platform track page scroll depth? Our previous approach involved analyzing where users typically leave the page. For instance, if a user exits after scrolling to the first screen, it suggests an issue with the initial content. To address this, we conducted interviews with target users to understand whether they could comprehend the content on the first screen, whether it captured their interest, and other related insights.
@gin_6078 yes there is a plugin I use called lucky orange! I will study screen recordings as traffic moves through the site. Did you actually reach out to users who bounced from your page? How did you get their details?
@emikes919 We are not focusing on those who have left but rather on the users who have stayed or their friends (who fit the user profile) to directly ask for insights.
@gin_6078 got it! im sure that was super valuable. keep grinding!
Absolutely feel your pain. One approach that helped me a ton was surveying the few visitors who didn't bounce. I's ask, "What almost made you leave this page?" Weirdly, those answers gave me insight into what probably caused bounces for others.
@david_grunwald1 very interesting. How did you survey them? Specifically... like how did you get their info, how did you reach out, etc? What tools?
Croct
@david_grunwald1 loved the approach!
Croct
I've been there and struggled with conversions for several years. Then I decided to dive deeper into not only AB testing but also personalization.
Before everything, a disclaimer: I'm the founder of Croct, an experimentation platform.
Typically, the primary reason users bounce is that the landing page message fails to resonate with what they were expecting in the first place. In your case, if you're testing different messages on ads (which is usual for companies trying to find market fit), it's essential to ensure the website messaging is consistent with the ad. Otherwise, they'll click on it expecting X and land on a website talking about Y.
After founding Croct, we've seen this pattern repeat over and over again among our customers. This may help you brainstorm some ideas, and this paybook shows how easy it is to apply them. Hope it helps!
@amorimj hey I looked at your website and blog posts, nice job! I think a youtube video version of this post would crush
Croct
@emikes919, thank you! Nice to know it is helpful :)
I will definitely work on a video format for this content, I appreciate the feedback!
Traffic is probably my biggest problem 😅
I'd reckon you would need at least 200 visitors to the same page to make it statistically relevant. Anything less than that and you are better off just tracking one patch at the time.
@dg_ what are you doing to drive traffic? any experience with paid ads?
@emikes919, I'm not at the point to pay for ads yet. I need ot make sure my messaging is resonate first, and I am doing that "by hand", meaning cold outreach, video chats, etc...
@dg_ I don't mean to be presumptuous about what you've got going on (financially or with your business), and I'll speak freely here.
I think this is a mistake founders make all the time. Paid ads are probably the most effective tool for testing messaging, and doing it quickly.
Just reacting to the language in your comment about being "at the point to pay", this reads like you view paid ads as expensive.
I see this as the wrong way to think about it.
Not only can you run paid ads to get hard data about how your message is resonating with huge reach very cheaply, but you can do it FAST.
And speed is where I think a lot of entrepreneurs misjudge things. If you can validating messaging quickly, no only do you save time outright, but you DERISK all future time and money spent on your project because you're acting based on a lot of hard data sooner. Everything you do going forward will be more effective.
I'm running some ads right now to test hooks for a broader ad campaign that I plan to scale, and the outperforming hooks can be used across all other channels too (landing page, content, emails, etc.) See screen shot (this is just a small portion of this test campaign). Basically spent no money and I have a pretty high degree of confidence in which language drives the most clicks.
Happy to chat with you off line about this if helpful.