Latest / The Bootstrapped Tech Founder with Fexingo: Profitable Software Companies Without VC / How Bootstrapped Founders Use Heatmaps Without A/B Testing
Transcript
- Lucas: Luna, I want to talk about something that sounds a bit counterintuitive for a bootstrapped founder: why you might not need A/B testing. Luna: Okay, I'll bite. A/B testing is practically gospel in growth circles. What's the alternative? Lucas: Heatmaps and session recordings. The argument is simple: when you're a team of one or two, you don't have enough traffic for statistically significant A/B tests. But you do have enough users to watch what they actually do. Luna: So instead of running experiments that take weeks to reach significance, you just... watch people? Lucas: Exactly. And it's surprisingly effective. I came across a case recently from a solo founder who runs a project management tool for remote teams. He had a free trial conversion rate around 8 percent, which he thought was normal. Then he installed a session recording tool and watched twenty sign-ups navigate his app. Luna: What did he find? Lucas: Almost every user who didn't convert spent the first five minutes lost. They clicked on non-interactive elements, ignored the onboarding checklist, and bounced to random help articles. He realized his tutorial video was autoplaying but muted, and users didn't know to enable sound. So he changed the onboarding to a simple walkthrough with clear next buttons. Luna: And? Lucas: Conversion went from 8 percent to just over 10 percent in two weeks. No A/B test, no developer hours on feature flags. Just watching recordings and making one change. Luna: That's a 25 percent lift from watching recordings. I'm sold. But what about the tools? There's Hotjar, Lucky Orange, Crazy Egg—they all have free or cheap tiers. Lucas: Right. And most bootstrapped founders can get meaningful data from the free plan of any of them. The key is knowing what to look for. I think there are three patterns that matter most. Luna: Let's hear them. Lucas: First: the rage click. A user clicks repeatedly on something that isn't a button. That's a design failure screaming for attention. Second: the dead end. A user reaches a page, scrolls slowly, then leaves. That page needs better hooks or clearer value. Third: the hidden feature. You built something great, but nobody clicks on it because it's buried or poorly labeled. Luna: I've definitely seen that last one in my own work. We had a feature that let users export reports as PDFs, but the icon was a tiny download arrow next to the chart. We moved it to a prominent button labeled 'Download Report' and usage tripled. Lucas: Exactly. And that's the kind of fix you could make in an afternoon without any testing infrastructure. Now, I'm not saying A/B testing is worthless. For high-traffic sites—say, an e-commerce store with tens of thousands of visitors—it's the gold standard. But for a bootstrapped SaaS with a few hundred trial users per month, the math doesn't work. Luna: Because you'd need thousands of visitors to detect a meaningful difference, right? Lucas: Right. Statistical significance at 5 percent lift with 80 percent power requires somewhere around 3,000 users per variant for a binary metric like trial conversion. Most bootstrapped apps don't have that. So you spend two weeks setting up a test, wait another month for data, and then the result is inconclusive. Luna: While heatmaps give you almost immediate insight. I like that. Lucas: And there's another angle: qualitative data often surfaces problems you didn't even know to test for. A/B testing can only compare variations you imagine. Session recordings show you what you're not thinking about. Luna: Like users trying to click on a static image because they think it's a button. Lucas: Exactly. Or users filling out a form but getting stuck on a field that's ambiguous. One founder I know noticed that in his checkout flow, users were entering their phone number with dashes and country codes, and the validator rejected it. He changed the input to accept any format, and abandoned cart dropped 15 percent. Luna: That's a huge improvement from one recording session. Lucas: Which brings me to a practical framework. I recommend bootstrapped founders do a weekly review: watch five session recordings, note three pain points, and fix one of them before the next week. That's a 30-minute habit that compounds quickly. Luna: And you don't need any special tools beyond a heatmap tool with a free tier. Lucas: Right. And if you're really early—like pre-launch—you can even record your own friends using a prototype and watch their sessions. It's the same principle: observe, note, fix. Luna: Okay, but let's push back on this a bit. Can heatmaps really replace A/B testing for pricing pages or checkout flows? Those are high-stakes pages where small changes can have big revenue impact. Lucas: Fair point. For pricing pages, I'd still use a combination. Heatmaps can show you where users' eyes go and what they click. But if you're changing pricing tiers, you might want an A/B test to measure revenue per visitor. The thing is, you can often get 80 percent of the way with heatmaps alone. Luna: So heatmaps first, then A/B test only if the change is high-risk or high-impact. Lucas: Exactly. And by the time you have enough traffic to run a proper A/B test, you'll already have learned so much from watching that you'll have better hypotheses. Luna: I want to circle back to the tools. Hotjar is probably the most well-known, but there's also FullStory and Microsoft Clarity, which is free. Lucas: Clarity is fantastic for bootstrapped founders because it's truly free—no time limits, no session caps. The trade-off is less advanced filtering, but for basic heatmaps and recordings, it's more than enough. Luna: And Crazy Egg has a very visual heatmap overlay that's great for landing page analysis. Lucas: Yeah. So the barrier to entry is essentially zero. The real constraint is actually your attention. You have to make time to watch the recordings. Luna: Honestly, if today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. Keeps the show ad-free and focused on practical, zero-budget advice like this. Lucas: Yeah, that's exactly it. No pressure, but if you found that heatmap approach useful, a coffee's worth a lot to us. And we're back — so let's talk about a mistake I see founders make with heatmaps. Luna: What's that? Lucas: Over-interpreting the first ten recordings. You'll see one user do something weird and think it's a universal pattern. You need at least 30 to 50 sessions before you trust a pattern. Luna: So sample size matters here too, just like A/B testing. Lucas: Right. But with recordings, you don't need statistical tests—you just need to see the same behavior three or four times. If three out of thirty users rage-click the same spot, that's a problem. You don't need a p-value to know that. Luna: I like that. It's more intuitive. Lucas: Another mistake is watching recordings at double speed. You'll miss subtle hesitations. A user hovering over a button for three seconds before clicking is a sign of confusion. At double speed, that's a blur. Luna: Good tip. I've definitely done that. Lucas: One more thing: use filters. If you're optimizing for conversion, only watch sessions of users who didn't convert. If you're optimizing for retention, watch sessions of users who came back. Don't mix them. Luna: That makes sense. Different goals, different behaviors. Lucas: Alright, let's wrap with a concrete action plan. If you're a bootstrapped founder listening, here's what I'd do this week. Pick one page that matters—pricing, sign-up, or onboarding. Watch 20 sessions using a free tool like Clarity. List three usability issues. Fix the easiest one. That's it. Luna: And if you do that every week for a month, you'll have a radically better product. Lucas: Exactly. And you'll never have to run a pointless A/B test again.