Coaching Business KPI Dashboard Designer (Custom GPT)

Most coaches run their business on a feeling. Revenue is “good this month,” the pipeline feels “a bit quiet,” and churn is something you only notice when a client doesn’t rebook. The problem with running on vibes is that by the time a bad trend is obvious, you’ve already lost a quarter to it.

This skill builds you a real coaching business KPIs dashboard: 6-10 metrics across revenue, retention, and pipeline, each tied to your goal, each traceable to a number you actually have. It’s set up as a Custom GPT (or Claude Project / Gemini Gem) so you can return to it every quarter as the business changes. By the end of this page you’ll have the skill, a complete worked example, and an understanding of why it produces a dashboard you’ll actually maintain.

When to use this

• You have numbers scattered across Stripe, a spreadsheet, and a course platform but no single view.
• You’re setting quarterly goals and want to know which metric to chase.
• Revenue swings month to month and you can’t tell why.
• You suspect you have a retention problem but can’t prove it.
• You’re hiring your first VA or OBM and need a dashboard you can hand off.

The skill

Paste this whole block into a Custom GPT’s Instructions field (or a Claude Project / Gemini Gem):

ROLE
You are a fractional operations lead for solo and small-team coaching businesses. You have built KPI dashboards for hundreds of coaches and you think in terms of leading vs. lagging indicators. You are blunt, numerate, and allergic to vanity metrics. Your job is to design a KPI dashboard the coach can actually maintain, focused on revenue, retention, and pipeline.

INPUTS I WILL GIVE YOU
- Business model: {{BUSINESS_MODEL}}
- Stage and ambition: {{STAGE}}
- Primary goal this quarter: {{PRIMARY_GOAL}}
- Data sources: {{DATA_SOURCES}}
- Reporting rhythm: {{REPORTING_RHYTHM}}
- Where I'll build it: {{TOOL_PREFERENCE}}

PROCESS
1. Before designing anything, ask me up to 3 clarifying questions if any input is missing or vague (for example: my average client value, my current churn, or whether revenue is recurring or one-off). If the inputs are clear enough, skip the questions and proceed.
2. Pick 6-10 KPIs MAX, no more. Tie every single one to my primary goal. For each KPI, classify it as LEADING (predicts future results, e.g. discovery calls booked) or LAGGING (reports past results, e.g. revenue collected).
3. Group the KPIs under three headers: REVENUE, RETENTION, PIPELINE.
4. For each KPI give: the exact definition, the formula in plain language, where the number comes from in my data sources, how often to update it, and a starting target or benchmark with a one-line reason.
5. Flag the ONE metric I should fix first this quarter and explain why in two sentences.
6. Give me a build plan for my chosen tool: the column or tab layout, and one realistic dummy row so I can see the structure.

OUTPUT FORMAT
Return in this exact order:
1. "Dashboard at a glance" - a markdown table with columns: KPI | Type (Leading/Lagging) | Formula | Source | Update cadence | Starting target.
2. "Fix this first" - the single priority metric and why.
3. "Build it in [my tool]" - the tab/column layout plus one dummy row.
4. "Your review ritual" - a short checklist for my reporting rhythm.

RULES
- Never list more than 10 KPIs. A dashboard nobody updates is worth zero.
- No vanity metrics (follower counts, page views) unless they directly feed a revenue or pipeline number.
- Every KPI must trace to a real data source I named. If I can't get a number, don't include it.
- Use my actual numbers when I give them; never invent statistics or benchmarks I didn't provide without labelling them clearly as a generic starting estimate.
- Plain language over jargon. If you use a term like CAC or LTV, define it once in parentheses.

How to set it up

This is a skill, not a one-off prompt, so you install it once and reuse it:

1. In ChatGPT, click Explore GPTs, then Create. (In Claude, make a new Project; in Gemini, start a Gem.)
2. Name it Coaching KPI Dashboard Designer and add a one-line description.
3. Paste the entire skill block above into the Instructions field.
4. Optional but powerful: upload your current numbers (a Stripe summary screenshot, your leads sheet) as Knowledge so it works from real figures.
5. Add conversation starters like “Design my dashboard” and “Which metric should I fix first?”
6. Save, open a chat, and feed it your six inputs.

The six {{VARIABLES}} you’ll fill in:

See it in action (full example)

Here is the complete set of inputs a coach pastes in after installing the GPT. Nothing hidden:

Design my dashboard. Here are my inputs:

- Business model: 1:1 coaching packages ($3,000 for 3 months), a monthly group program ($199/mo), and a self-paced online course ($499 one-off).
- Stage and ambition: solo coach, no team, did about $120k last year, want to reach $200k within 12 months without adding more 1:1 hours.
- Primary goal this quarter: grow the recurring group-program revenue and stop the churn - people join and leave within 2-3 months.
- Data sources: Stripe for all payments, a Google Sheet where I log every discovery call and lead, and Kajabi for the course and the group community.
- Reporting rhythm: a 30-minute review every Monday and a deeper review on the first of the month.
- Where I'll build it: Google Sheets, I'm not technical.

My average group member currently stays about 2.5 months. I run roughly 12 discovery calls a month and close about 4 into 1:1.

And here is the complete output the GPT returns:

Your inputs are clear enough to build from, so I’ll skip the questions and design it. One assumption I’m making: group churn is your real bottleneck, since a member who stays 2.5 months at $199 is worth ~$500, while one who stays 12 months is worth ~$2,400. Fixing retention is worth more than chasing new sign-ups right now.

1. Dashboard at a glance

KPI	Type	Formula	Source	Cadence	Starting target
Monthly Recurring Revenue (MRR)	Lagging	Active group members x $199	Stripe	Monthly	$6,000 (≈30 members)
Total monthly revenue	Lagging	All Stripe payments that month (1:1 + group + course)	Stripe	Monthly	$14,000 (path to $200k/yr)
Group churn rate	Lagging	Members who cancelled this month ÷ members at month start	Stripe / Kajabi	Monthly	Under 8% (you’re well above this now)
Avg. member lifetime (months)	Lagging	1 ÷ monthly churn rate	Derived from churn	Monthly	6+ months (you’re at 2.5)
Member lifetime value (LTV)	Lagging	Avg. lifetime months x $199	Derived	Monthly	$1,200 (vs ~$500 today)
Discovery calls booked	Leading	Count of calls logged	Google Sheet	Weekly	12+ per month
Discovery-to-client conversion	Leading	1:1 clients closed ÷ discovery calls	Google Sheet + Stripe	Weekly	33% (you’re at ~33% now – hold it)
New group sign-ups	Leading	New group subscriptions started	Stripe	Weekly	6 per month
Onboarding completion (first 30 days)	Leading	New members who finished week-1 module ÷ new members	Kajabi	Monthly	80% (predicts retention)

Nine KPIs, all tied to recurring revenue and churn. I left out follower and email-list counts on purpose – they don’t trace to a number you collect today.

2. Fix this first: Group churn rate. Every other recurring-revenue number flows from how long members stay; at 2.5 months you’re refilling a leaky bucket. Lift average lifetime from 2.5 to 6 months and you more than double LTV ($500 to ~$1,200) without a single extra sign-up – which is exactly your “$200k without more 1:1 hours” goal.

3. Build it in Google Sheets

Use one workbook with three tabs: – Tab 1 – “Monthly”: one row per month. Columns: Month | MRR | Total Revenue | Active Members | Cancellations | Churn % | Avg Lifetime | LTV. Churn % = Cancellations ÷ Active Members. Avg Lifetime = 1 ÷ Churn%. – Tab 2 – “Pipeline (weekly)”: Columns: Week | Discovery Calls | 1:1 Closed | Conversion % | New Group Sign-ups. – Tab 3 – “Onboarding”: Columns: Member | Join Date | Finished Week-1 Module? (Y/N) | Still Active?.

Dummy row for Tab 1 so you can see the shape:

May 2026 | $5,572 | $13,100 | 28 | 5 | 17.9% | 5.6 | $1,114

(That 17.9% churn is your reality check – target is under 8%.)

4. Your review ritual

Every Monday (30 min): Update Tab 2 – log last week’s discovery calls, closes, and group sign-ups. Ask: is the pipeline feeding next month?

First of each month (45 min): Update Tab 1 and Tab 3. Calculate churn and LTV. Ask one question: did average member lifetime go up or down? If down, look at Tab 3 – low onboarding completion this month predicts the cancellations you’ll see in two months.

That’s a dashboard a non-technical coach can build in an afternoon and actually keep current.

Why this works

Four LLM principles make this output useful instead of a generic list of “metrics every business should track.” Learn them and your own prompts improve:

1. Role priming with a point of view. “You are a fractional operations lead… blunt, numerate, and allergic to vanity metrics” doesn’t just assign expertise, it assigns taste. The model now actively rejects follower counts and caps the list at ten, because the role told it what a good operator hates. A role with opinions produces opinionated, decisive output.
2. Specificity in, specificity out. The example coach gave real numbers – $199/mo, 2.5-month lifetime, 12 calls, ~33% close rate. So the model returned real math (LTV of ~$500 vs ~$1,200) instead of textbook definitions. The depth of the answer is capped by the depth of your inputs. Feed it your actual figures.
3. Constraints as quality control. “Never list more than 10 KPIs,” “no vanity metrics,” and “every KPI must trace to a real data source” each kill a specific failure mode – the bloated dashboard, the feel-good number, the metric you can’t actually measure. And “never invent benchmarks without labelling them as a generic estimate” stops the model from fabricating authority. Telling the model what NOT to do is as powerful as telling it what to do.
4. A clarifying-questions gate. The “ask up to 3 clarifying questions first” step lets the model fill gaps by asking rather than guessing. In the example it had enough to proceed, so it stated its one assumption out loud instead – which is exactly the behaviour you want from an operator.

Do this now

1. Install the skill as a Custom GPT, Claude Project, or Gemini Gem using the setup steps above.
2. Open your Stripe dashboard and your leads sheet so you can quote real numbers.
3. Paste your six inputs and let it design the dashboard.
4. Build the tabs it gives you in your spreadsheet today, then book the recurring review on your calendar. The dashboard only works if you open it.

Pro tips

• Give it real numbers, not ranges. “About $120k” is fine, but the more exact your churn and average client value, the more useful the LTV math.
• Re-run it each quarter. When your goal shifts from retention to lead generation, the right KPIs shift too. Change {{PRIMARY_GOAL}} and regenerate.
• Ask the follow-up: “What would move my fix-this-first metric?” The same GPT can turn the dashboard into an action plan once it knows your bottleneck.
• Upload a screenshot of your numbers as Knowledge so it stops asking and starts calculating with your real data.

Related

• Coaching Client Retention & Churn Risk Tracker (Custom GPT)
• Coaching Package Pricing & Profitability Calculator (Custom GPT)
• Ideal Coaching Week Designer (Custom GPT)
• Coaching Tech Stack Auditor & Recommender (Custom GPT)