These prompts help teams use AI to analyze heatmap data faster and more rigorously. Replace [bracketed placeholders] with your specifics before pasting into ChatGPT, Claude, or another LLM.
Prompt 1: Interpret a click heatmap
Here is a click heatmap export for [page name on website].
Page goal: [the specific action the page is supposed to drive — e.g., "click the Start Free Trial CTA"]
Click data (top 10 clicked elements):
[Element 1]: [click count] clicks, [%] of total clicks
[Element 2]: [click count] clicks, [%] of total clicks
...
Context: [traffic volume, segment if any, key page sections]
Please:
1. Identify whether the primary CTA is receiving the expected share of clicks for a converting page (typical benchmark: 5–15% of visitors)
2. Flag any element with a high click count that is not interactive (potential dead clicks) and suggest what users likely expected
3. Identify elements that should have received clicks but did not, and propose visibility or labeling fixes
4. Suggest 2–3 specific design changes to improve the click distribution toward the page goal
5. Recommend which session replays to watch to validate the hypothesesPrompt 2: Analyze a scroll heatmap and find false bottoms
Here is the scroll depth data for [page name].
Scroll depth distribution:
- 100% of visitors viewed: [section name, e.g., hero]
- 75% reached: [section]
- 50% reached: [section]
- 25% reached: [section]
- 10% reached: [section]
Page goal: [e.g., "users complete the sign-up form at 80% scroll depth"]
Page length: [pixels or sections]
Please:
1. Identify the steepest drop-off point (where the largest percentage of visitors stops scrolling)
2. List the page elements that sit below the 50% scroll line and assess whether any of them are critical to the page goal
3. Flag any layout patterns that may create a "false bottom" (large whitespace, horizontal divider, full-width image) at the drop-off point
4. Recommend which sections to move higher and which to remove or condense
5. Estimate the conversion impact if 20% more visitors reached the goal sectionPrompt 3: Compare segments on the same page
I have heatmap data for [page name] split by [segment dimension — e.g., desktop vs mobile, paid vs organic].
Segment A ([name]): [traffic volume]
Top 5 clicked elements: [list with click share]
Scroll-to-50% point: [section name]
Notable patterns: [dead clicks, rage clicks, anomalies]
Segment B ([name]): [traffic volume]
Top 5 clicked elements: [list with click share]
Scroll-to-50% point: [section name]
Notable patterns: [dead clicks, rage clicks, anomalies]
Please:
1. Identify the largest behavioral differences between the two segments
2. Suggest which differences are explained by the medium (e.g., thumb-reach on mobile) vs by user intent (e.g., paid users have a different goal)
3. Recommend whether the page needs separate optimizations for each segment or a single design that serves both
4. List the specific design changes for each segment, prioritized by expected impactPrompt 4: Generate a heatmap audit report
I ran heatmap analysis on [number] pages of [website name]. Please produce a structured audit report from the findings below.
For each page, I have:
- Page name and primary goal
- Top click pattern and any dead-click clusters
- Scroll depth at 50% and at the goal section
- Notable rage clicks or anomalies
- Segment differences (if any)
Page 1: [findings]
Page 2: [findings]
...
Please structure the report as:
1. Executive summary: 3 patterns that repeat across pages
2. Per-page findings: one section per page with What / So What / Now What
3. Cross-cutting recommendations: design system or template changes that would fix multiple pages at once
4. Prioritized action list: the top 5 changes ranked by expected impact and implementation effort
5. Measurement plan: which metrics to watch after each change ships