HomeCasa KPI definitions and dashboard spec: formulas, sources and owners
Define your KPIs with no ambiguity: a metric tree, a definition card per KPI with exact formula, source and owner, and a dashboard spec your data team can build straight away. Designed for ChatGPT, Claude, Gemini and Microsoft Copilot. It includes 5 customizable variables and you can copy it for free in one click.
Advanced · Text · Pending editorial review
Prompt
Act as a head of business analytics who turns goals into measurable metrics and specs that a data team can build without guessing.
Context:
- Type of business or team: {{business}}
- Main goal for the next period: {{goal}}
- Who will use the dashboard: {{audience}}
- Available data sources: {{sources}}
- Refresh frequency: {{frequency}}
If the goal is vague or an essential source is missing, ask me before defining metrics.
Deliverables:
1. A metric tree: one primary metric tied to the goal, 3-5 driver metrics and 2-3 guardrail metrics that must not get worse.
2. A KPI card for each metric: name, one-sentence definition, exact formula (numerator and denominator), filters and exclusions, grain, source and field, owner, frequency, target [FILL IN], leading or lagging, and how it could be gamed or misread.
3. Edge cases to decide (refunds, test accounts, time zones, currency) marked [FILL IN: decision].
4. Dashboard spec: a top row of headline KPIs compared with target and prior period, trend charts, breakdowns and a detail table; the recommended chart type for each KPI and why.
5. Data quality checks and alerts (nulls, sudden jumps, late data loads).
6. Metrics to leave out because they are vanity or redundant.
7. A review ritual: who looks at what, when, and which decisions each KPI informs.
Constraints:
- Do not invent benchmarks or targets: use [FILL IN].
- Use definitions that can be calculated from the listed sources; if that is not possible, say so.
- Plain enough for business readers and precise enough for the technical team.
Format: the tree as a nested list, a table of KPI cards and an outline of the dashboard by section.How to use it
Make the goal specific and list every real source (CRM, web analytics, accounting). Share the KPI cards with each owner to settle the edge cases before anyone builds the dashboard.
Usage example
Business: online skincare store · Goal: increase repeat purchases · Audience: leadership · Sources: Shopify, Google Analytics 4, cost sheet · Frequency: weekly.
Sample output
1. Metric tree
- Primary: 90-day repeat purchase rate
- Drivers: % of first-time buyers subscribed to email · post-purchase flow conversion (Klaviyo) · median days to second order · % of customers buying a refillable product
- Guardrails: contribution margin per order · % of orders with a discount · email unsubscribe rate
2. KPI cards (excerpt)
| Field | 90-day repeat rate | Contribution margin per order |
|---|---|---|
| Definition | % of new customers in a weekly cohort who place a 2nd order within 90 days | What each order leaves after variable costs |
| Formula | customers with 2nd order ≤ 90 days / customers with 1st order in week W | (net revenue − product cost − shipping − fees) / orders |
| Filters | Exclude test orders and fully refunded orders | Exclude cancelled orders |
| Grain · source | Weekly cohort · Shopify orders (customer_id, created_at) | Order · Shopify + cost sheet |
| Owner · target | CRM lead · [FILL IN] | Finance · [FILL IN] |
| Type | Lagging (cohort matures after 90 days) | Coincident |
| How it can mislead | Pushed up by heavy discounting | Hides free-shipping costs |
3. Edge cases to decide
- Does a refunded order count as a repeat? [FILL IN: decision]
- Time zone used to cut the week: [FILL IN: decision]
4. Dashboard (top row)
Repeat rate (latest mature cohort) vs target and prior cohort · margin per order · % discounted orders. Below: repeat curve by cohort (line chart) and breakdown by first product purchased (bar chart). …
Key facts
- Category
- Data management · Reports
- Output type
- Text
- Level
- Advanced
- Designed for
- ChatGPT, Claude, Gemini and Microsoft Copilot
- Verified in
- Pending editorial verification
- Variables
- Business or team, Main goal, Audience, Data sources and Refresh frequency
- Language
- English
- Price
- Free, no sign-up
- Updated
- Nov 7, 2026
Frequently asked questions
Why include guardrail metrics?
Because optimizing one metric can quietly hurt another, for example growing sales through discounts that wipe out margin. Guardrails make that visible.
How is this different from designing the dashboard UI?
This prompt focuses on what to measure and how to calculate it unambiguously. Visual design comes next and is far easier once the definitions are settled.
What is this prompt for?
Define your KPIs with no ambiguity: a metric tree, a definition card per KPI with exact formula, source and owner, and a dashboard spec your data team can build straight away.
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 (Business or team, Main goal, Audience, Data sources and Refresh frequency). 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.