A client we picked up last quarter handed us their analytics stack on the first call. They were paying for Hotjar at the $400-a-month tier, FullStory at roughly $1,000, GA4, and a Mixpanel seat that nobody had logged into in five months. Total bill was nearly $20,000 a year. When we asked what they'd shipped because of the data, the marketing lead thought for a moment and said "we changed a button color once."

That's the common case. Teams buy the tools. The tools collect data. The data sits there.

Heatmaps and session replays are both useful. They are useful for different reasons, in different moments, and the failure mode of most teams is that they buy both and use neither.

What each tool actually shows you

A heatmap is an aggregate. You get a click map for a page that tells you that 38% of visitors clicked the hero CTA, 12% clicked the secondary CTA, and 7% clicked a piece of decorative text that nobody designed to be clickable. You get scroll depth showing that 60% of users never reach the second section. You get rage clicks aggregated across a thousand sessions on a button that isn't a button.

It's quantitative. It tells you what's true on average across many users. It's bad at telling you why.

A session replay is a story. You watch one user, from the moment they arrived to the moment they left. You see them hover on the price for a long time, scroll back up, scroll to the testimonials, scroll back to the price, and close the tab. A heatmap would tell you that page had a 78% bounce rate. A replay tells you exactly what hesitation looks like on that page.

It's qualitative. Five sessions can teach you more about why something isn't working than a million aggregate clicks. It's bad at telling you whether what you observed is typical.

When each one is worth paying for

Heatmaps are worth their cost on sites where you're optimizing pages that already have meaningful traffic. If you have 5,000+ sessions a week on a landing page, the heatmap is telling you something statistically interesting and you can act on it. Move a button. Change copy. Re-run. The data refreshes fast and the changes are testable.

Session replays are worth their cost when you have a specific question. Why are users dropping off at step three of a five-step funnel? Why does the cart abandonment rate spike on iPhone Safari? What does the experience actually look like for someone arriving from a paid Google ad versus organic search? You can answer those questions by watching ten replays. You cannot answer them with heatmaps.

The trap is buying both, looking at neither, and feeling like you're being data-driven because you check a box on a vendor checklist.

The two-week rhythm we recommend

We tell clients to set a Friday afternoon recurring calendar event. Two weeks. Two hours. The marketing person watches ten session replays of users who completed the desired action and ten who didn't. They write down three observations. They do nothing else with the tool in between those two-hour blocks.

That's it. Ten and ten, every two weeks. Twenty replays, forty over a month. You will see patterns. You will form hypotheses. The hypotheses go into a list. The list drives the next round of A/B tests on the pages where you have enough traffic for heatmaps to be useful.

The two tools work together this way. Replays generate ideas. Heatmaps validate them. Most teams skip the replay-watching part because it feels less productive than looking at a number, and then they wonder why their A/B tests come up flat.

The specific tools, briefly

Hotjar is fine for SMBs. The free tier is genuinely useful up to 35 sessions per day, and the paid plans top out at prices small businesses can absorb. The heatmaps are sufficient. The replays are slightly clunky to filter but good enough.

FullStory is overkill for most of our clients. It's a great product if you have a team of CX analysts whose full-time job is digging through user behavior. If you're a small business with a marketing manager who also runs the email program, you will not extract $1,000 a month of value from it.

Microsoft Clarity is free, doesn't sample sessions, and is roughly 80% as useful as Hotjar for our purposes. We recommend it constantly to clients who balked at paying. The downside is it integrates with fewer other tools and the data export is clunky, but for a small business that just wants to watch replays, it's the right starting point.

The point

If you can only buy one, buy heatmaps if you have meaningful traffic and a clear page you're optimizing. Buy session replays if you have a funnel with a leak you can't explain. If your team won't carve out two hours every two weeks to actually watch replays, don't buy the replay product — it'll be a $12,000-a-year line item that produces nothing.

The "we changed a button color once" client cancelled FullStory, kept Hotjar on the smallest paid tier, and replaced two of the seats with Clarity. The marketing lead now does the two-hour Friday session. Three months in, they've shipped four changes off observations from replays and seen lifts on three of them. The tool didn't change. The discipline did.