HomeCasa Prioritized CRO hypothesis backlog for your online store
Turn funnel data and customer feedback into a CRO hypothesis backlog: diagnosis, 10-15 well-formed hypotheses, ICE scoring, test feasibility checks and experiment briefs. Designed for ChatGPT, Claude, Gemini and Microsoft Copilot. It includes 4 customizable variables and you can copy it for free in one click.
Advanced · Text · Pending editorial review
Prompt
Act as an e-commerce conversion rate optimization (CRO) specialist with solid experimentation and basic statistics skills.
Context:
- Store and main product: {{store}}
- Funnel data (sessions, product views, add to cart, checkout, purchase; by device if available): {{funnel}}
- Qualitative evidence (reviews, tickets, surveys, session recordings, heatmaps): {{evidence}}
- Monthly traffic and monthly orders: {{traffic}}
Deliverables:
1. Diagnosis: at which funnel step and on which device most people drop off, with percentages calculated from my data.
2. 10-15 hypotheses in this format: "Because we observed [evidence], we believe that changing [element] for [segment] will cause [effect]. We will measure it with [metric]." Each must cite the specific evidence it relies on.
3. An ICE prioritization table (Impact, Confidence, Ease, 1 to 10) with a short justification for each score.
4. Feasibility: given {{traffic}}, estimate roughly whether an A/B test can detect a reasonable effect within 2-6 weeks. If not, recommend alternatives (ship and monitor, usability testing with 5 users, testing on a higher-traffic page).
5. An experiment brief for the top 3: variant description, primary metric, guardrail metrics (for example, returns or average order value), estimated duration, stopping rule and risks.
6. A learning log template to document every test.
Constraints:
- Do not invent industry benchmarks or promise uplifts. If you use an assumption in a calculation, state it explicitly.
- Do not suggest dark patterns (fake urgency, pre-checked boxes, costs hidden until the last step).
- If the data has gaps, mark them [FILL IN] and ask me before prioritizing.
Format: a short diagnosis, a hypothesis table with ICE sorted from highest to lowest, and experiment briefs under headings.How to use it
- Export the last 30-90 days of funnel data from your analytics. 2. Paste real snippets from reviews or tickets as evidence. 3. Run one experiment at a time and log the result in the template, win or lose.
Usage example
Store: running shoes · Funnel: 40,000 sessions, 2.1% add to cart on mobile vs 4.5% on desktop · Evidence: tickets about sizing doubts · Traffic: 40,000 sessions and 520 orders a month.
Sample output
1. Diagnosis
| Step | Sessions | Step conversion |
|---|---|---|
| Product view | 10,800 (18%) | — |
| Add to cart | 1,920 (3.2%) | 17.8% |
| Start checkout | 1,140 (1.9%) | 59.4% |
| Purchase | 540 (0.9%) | 47.4% |
The biggest leak is product page to cart: 82% of product viewers don't add to cart. Mobile purchase rate is 0.6%, a third below the average. [FILL IN: funnel by device to see which step drops on mobile].
2. Hypotheses (excerpt)
H1. Because we observed drop-offs at size selection in recordings, we believe that a visual size picker showing bed dimensions for mobile users will increase add-to-cart rate. We will measure it with add to cart / product views on mobile.
H2. Because we observed reviews asking to compare firmness, we believe that a firmness comparison table on the product page will increase add-to-cart rate…
3. ICE prioritization
| Hypothesis | I | C | E | ICE |
|---|---|---|---|---|
| H1 Size picker | 8 | 7 | 6 | 7.0 |
| H2 Firmness comparison | 7 | 7 | 7 | 7.0 |
| H3 Delivery time on product page | 6 | 6 | 9 | 7.0 |
4. Feasibility
At 540 orders a month, detecting a +10% lift in purchases would need roughly 175,000 sessions per variant (assumption: 80% power, 5% significance). Not feasible in 6 weeks. Testing add to cart on the ~10,800 monthly product views is…
Key facts
- Category
- E-commerce · Conversion optimization
- Output type
- Text
- Level
- Advanced
- Designed for
- ChatGPT, Claude, Gemini and Microsoft Copilot
- Verified in
- Pending editorial verification
- Variables
- Store and main product, Funnel data, Qualitative evidence and Monthly traffic and orders
- Language
- English
- Price
- Free, no sign-up
- Updated
- Nov 7, 2026
Frequently asked questions
What is ICE scoring?
A quick prioritization method: you score each hypothesis on potential Impact, your Confidence it will work and the Ease of implementing it. Sorting by the average helps you start with the most promising, cheapest ideas.
Do I need a lot of traffic to do CRO?
Reliable A/B tests do need volume. With low traffic, the prompt suggests alternatives such as usability tests or evidence-based changes that you ship and monitor.
What is this prompt for?
Turn funnel data and customer feedback into a CRO hypothesis backlog: diagnosis, 10-15 well-formed hypotheses, ICE scoring, test feasibility checks and experiment briefs.
Which AI tools does it work with?
It is designed for ChatGPT, Claude, Gemini and Microsoft Copilot. We have not verified it in these tools yet, so results may vary between them.
How do I customize the prompt?
Click “Customize” and fill in the variables (Store and main product, Funnel data, Qualitative evidence and Monthly traffic and orders). The text updates instantly and you can then copy it in one click. Values are stored only in your browser.
Do I need experience to use it?
It is advanced: it is meant for people with experience in the topic who can evaluate and adjust the result.
Is this prompt free?
Yes. Every HomeCasa Prompts prompt is free, with no sign-up or subscription. You can also use them in professional projects.
Can I trust the AI output?
Treat it as a high-quality draft: always check data, figures and important claims before publishing or making decisions.