Key takeaways
- The S-Tier CSM Scorecard has seven numbers: retention across the book, time to first win, refund and chargeback rate, renewal and ascension rate, program completion, case studies captured and referrals generated. Add response time and NPS.
- Every KPI needs a written formula and a clear time window, or two people will report two different numbers for the same month.
- Leading indicators (attendance, check-in replies, time to first win) tell you what retention and refunds will look like weeks before they happen.
- A client health score is a simple weighted sum of a few honest inputs, turned into green, yellow and red, and it only works if you act on red within a day or two.
- Founders read one page, not a dashboard. A weekly report with the numbers, the trend, the red clients and one ask is enough.
The client success KPIs that matter in a coaching business are the ones that show whether clients get results and stay: retention across the book, time to first win, refund and chargeback rate, renewal and ascension rate, program completion, case studies captured and referrals generated. Add response time to at-risk clients and NPS, track all of it in one simple sheet, and report it to the founder once a week.
Most customer success metrics you find online come from SaaS, where the question is whether people log in and use a product. In coaching and agencies the question is different. Did this person get the result they paid for, and will they stay, renew and tell their friends?
I'm Diego, founder of CSM Council. Through my agency, Scale With Fulfillment, I run fulfillment as a fractional COO inside coaching offers, some of them doing million-dollar months. This guide is how I want CSMs to measure their book: what to track, how to calculate it, and what to do when a number moves.
Why client success KPIs matter in coaching
Without numbers, a CSM gets judged on feelings. The founder hears about the one angry client in Slack and decides fulfillment is broken. Or nobody notices the quiet clients until the refund request lands.
KPIs fix three things:
- You see problems early. A drop in attendance this week is a refund request next month.
- You prove your value. Renewals, saves and case studies are revenue. Numbers make that visible to the founder.
- You stop arguing about opinions. "Clients seem unhappy" turns into "four clients are red, here is the plan for each."
That is the difference between a CSM who executes and an S-Tier CSM who drives revenue. The second one knows their numbers.
The S-Tier CSM Scorecard: 7 numbers plus 2
The S-Tier CSM Scorecard is the set of numbers we use in CSM Council to define a top-performing CSM. Here is the overview before we go through each one.
| KPI | Type | What it tells you |
|---|---|---|
| Retention across the book | Lagging | Are clients staying? |
| Time to first win | Leading | Are clients getting momentum fast? |
| Refund and chargeback rate | Lagging | Are we losing cash we already collected? |
| Renewal and ascension rate | Lagging | Do clients want more after it worked? |
| Program completion | Leading | Are clients doing the work? |
| Case studies captured | Lagging | Are results turning into proof? |
| Referrals generated | Lagging | Do clients trust us enough to send friends? |
| Response time | Leading | How fast do we act when something is off? |
| NPS | Leading | How do clients feel right now? |
One rule before the formulas: write each definition down and never change it quietly. If "active client" means something different in March than in February, your trend is useless.
1. Retention across the book
Definition: the share of clients who were active at the start of a period and are still active at the end.
Formula: (clients active at start minus clients who left during the period) divided by clients active at start, times 100.
New clients who joined during the period stay out of the calculation. Otherwise a big sales month hides churn.
What "left" means: refunded, cancelled a payment plan, stopped paying, or formally dropped out. Decide with your founder whether a client who simply went silent counts. I prefer to track silent clients separately as "at risk" rather than "left," so the number stays honest.
When it moves down: look at who left and when. If most leave in the first 30 days, it is an onboarding problem. If they leave in the middle, it is usually an engagement or belief problem. Our client retention guide goes deeper on fixes by stage.
2. Time to first win
Definition: the number of days from purchase to the client's first confirmed result.
Formula: date of first win minus purchase date, per client. Report the median for clients who started in the period.
The hard part is defining the "first win" for your program. It must be specific and observable. "Feels more confident" is not a win. "Launched the first ad," "booked the first sales call" or "lost the first two kilos" are wins.
When it moves up: clients are getting stuck early. Check onboarding call attendance, how fast access gets set up, and whether week one asks too much. This is what the Day-21 Activation Sprint framework in our training is about: getting every client to a first real win inside three weeks.
3. Refund and chargeback rate
Definition: the share of clients who got money back, either through a refund or a payment dispute.
Formula: (refunds plus chargebacks in the period) divided by new clients in the same period, times 100. Track refunds and chargebacks in separate columns too, because they mean different things. A refund is a conversation that ended badly. A chargeback often means there was no conversation at all.
When it moves up: read every refund reason and sort them into three buckets: wrong fit (a sales problem), no results yet (a delivery problem), and life events. Share the buckets with the founder. Most of the fixable ones live in the second bucket. The refund and churn guide walks through save conversations step by step.
4. Renewal and ascension rate
Definition: the share of eligible clients who renew the same program or move up to the next offer.
Formula: (clients who renewed or ascended) divided by (clients whose program ended or who were eligible for the next offer in the period), times 100.
"Eligible" matters. A client in week three is not a renewal candidate. Count only clients who reached the renewal window.
When it moves down: check whether those clients actually got results. Renewal is mostly decided months earlier, by time to first win and completion. Then check whether anyone had the conversation at all. Many CSMs avoid it because it feels like sales. More on that in renewals and upsells for coaching clients.
5. Program completion
Definition: the share of clients who complete the core milestones of the program on time.
Formula: clients who hit milestone X by the target week, divided by clients who reached that week, times 100. Track it per milestone, not just "finished the whole thing," so you see exactly where people drop.
You can add attendance here too: calls attended divided by calls offered, per client.
When it moves down: find the milestone where the drop happens. If 80% of a cohort hits milestone two and only half hit milestone three, milestone three is the problem. That is a fulfillment fix, and bringing it to the founder is exactly the kind of systems thinking coaches hire for.
6. Case studies captured
Definition: the number of testimonials and case studies collected in the period.
Formula: simple count, split into short testimonials (text or video) and full case studies (before, after, numbers, story).
When it moves down: you are probably not asking at the right moment. The best moment is right after a win, not at the end of the program. Put a "proof ask" step into your process every time a client posts a win.
7. Referrals generated
Definition: the number of new leads or clients that came from your current book.
Formula: count of referred leads, and separately, referred leads who bought. Ask sales to tag referrals in the CRM, or you will never see them.
When it moves down: referrals follow results and relationship. If both are fine, you are likely not asking. A simple ask after a win or a high NPS score is often enough.
8. Response time
Definition: how long it takes to contact a client after they get flagged at risk, or after they send a message that needs a real answer.
Formula: hours from flag (or message) to first meaningful response. Report the median, and the longest one.
When it moves up: your book is too big, your process is too manual, or you are spending time in the wrong places. Fix the process first, then talk to the founder about book size.
9. NPS
Definition: Net Promoter Score, based on the question "How likely are you to recommend this program to a friend?" on a 0 to 10 scale.
Formula: percentage of 9s and 10s (promoters) minus percentage of 0 to 6 (detractors). Sevens and eights count as passives.
How to use it: ask at fixed points, like day 30 and the end of the program. Read every comment. Call every detractor within two days. Ask every promoter for a testimonial or a referral.
When it moves down: treat it as an early warning, not a verdict. Small books make NPS jumpy, so look at the trend over several weeks and at the comments.
How to track it simply: one sheet
You do not need a customer success platform. Most coaching offers I work with run fine on one shared sheet, or a simple board in their CRM. Here is the structure.
Tab 1: Clients (one row per client)
| Column | Example |
|---|---|
| Client name | Sarah K. |
| Program and start date | Accelerator, Sep 2 |
| Goal (in their words) | First 3 paying clients |
| First win date | Sep 16 |
| Milestones hit | 3 of 6 |
| Calls attended / offered | 5 / 6 |
| Last reply to check-in | Oct 24 |
| Health score and color | 82, green |
| NPS (latest) | 9 |
| Status | Active / refunded / chargeback / renewed / ascended / left |
| Proof captured | Video testimonial, Oct 1 |
| Referrals | 1 lead |
| Next action | Ask for case study call |
Tab 2: Weekly scorecard (one row per week)
Formulas pull from Tab 1: retention, median time to first win, refund and chargeback rate, renewal and ascension rate, completion per milestone, proof count, referrals, median response time, NPS.
Your weekly tracking checklist:
- Update attendance and milestone columns
- Update "last reply" for every client
- Recalculate health scores and colors
- Log every refund, chargeback, renewal and ascension with a reason
- Log proof and referrals
- Copy this week's numbers into Tab 2
Thirty minutes a week, once it is set up. If your offer uses Skool, Slack and a CRM, pull attendance and replies from there, but keep the sheet as the one source of truth.
Become an S-Tier CSM.
Earn more and work with the top 1% of coaching offers & agencies.
Building a client health score
A client health score turns a few signals into one number per client, so you know who to call today. The Client Health Score is one of the frameworks we teach, and the simple version below is enough to start.
Step 1: pick 4 to 6 honest inputs
Choose signals you can actually see every week. For a coaching program, these usually work:
- Attendance: calls attended in the last two weeks
- Responsiveness: replied to the last check-in, and how fast
- Progress: milestones hit on time vs behind
- Activity: posts, questions or submissions in the community
- Sentiment: your read from the last conversation, plus latest NPS
- Payment: payment plan on time or failed
Leave out inputs you would have to guess. A health score built on guesses gives you confidence you should not have.
Step 2: give each input a weight
Weights reflect what predicts churn in your program. Start with your best judgment, then adjust after a few months by looking at who actually left.
Step 3: map the total to green, yellow and red
Worked example (for illustration only, not a benchmark):
| Input | Weight | How to score it (0 to 100) |
|---|---|---|
| Attendance | 25% | 100 if all calls attended, 50 if half, 0 if none |
| Responsiveness | 20% | 100 replied within 48h, 50 within a week, 0 no reply |
| Progress | 25% | 100 on track, 50 one milestone behind, 0 two or more behind |
| Activity | 10% | 100 active this week, 50 last two weeks, 0 none |
| Sentiment | 15% | 100 positive, 50 neutral or unclear, 0 negative |
| Payment | 5% | 100 on time, 0 failed or late |
In this example, a client who attended half the calls (50), replied within a week (50), is on track (100), was active last week (100), sounded neutral (50) and paid on time (100) scores:
(50 × 0.25) + (50 × 0.20) + (100 × 0.25) + (100 × 0.10) + (50 × 0.15) + (100 × 0.05) = 12.5 + 10 + 25 + 10 + 7.5 + 5 = 70.
Example color bands: 75 to 100 green, 50 to 74 yellow, below 50 red. So this client is yellow. Set your own bands, and check them against reality every month.
Step 4: attach an action to every color
A score with no action is decoration.
| Color | What it means | What you do |
|---|---|---|
| Green | On track | Celebrate wins, ask for proof, note renewal signals |
| Yellow | Drifting | Personal message or short Loom within 48 hours, agree on one next step |
| Red | At risk | Call within 24 hours, find the real reason, agree on a two-week plan, flag to the founder |
One override rule: a client who mentions a refund, a chargeback or "this isn't working" is red, no matter what the score says.
For the red-client call, a simple opener works:
"Hey Sarah, I noticed you missed the last two calls and I didn't want to just send another reminder. How are things really going with the program? What's getting in the way right now?"
Then listen. The first answer is rarely the real one.
Leading vs lagging indicators
This is the part most CSMs miss, and it changes how you spend your week.
Lagging indicators tell you what already happened: retention, refund and chargeback rate, renewal and ascension rate, case studies, referrals. By the time they move, it is too late to change that month.
Leading indicators tell you what is about to happen: attendance, check-in replies, time to first win, milestone completion, health score, response time, NPS.
| Leading indicator | Lagging result it predicts |
|---|---|
| Slow time to first win | Refunds in the first 30 to 60 days |
| Dropping attendance | Churn and failed payments |
| Milestone stalls | Low completion, low renewal rate |
| Slow response time | More reds turning into refunds |
| High NPS after a win | Testimonials and referrals |
The rule: manage the leading indicators every week, report the lagging ones to the founder. If you only look at retention, you are driving by looking in the rearview mirror.
Inside CSM Council: you get the S-Tier CSM Scorecard, the Client Health Score and the weekly founder report as ready-to-use templates, so you can track these numbers from day one. No one is guaranteed a role or a result, but you'll walk into every founder meeting with the numbers that matter. See what's inside the community.
Weekly reporting to the founder
Founders are busy. They will not open your dashboard. They will read one short message, every week, at the same time. Make it skimmable, honest and end with one clear ask.
Weekly report template:
Client Success: week of [date]
Book: [X] active clients. [X] green, [X] yellow, [X] red.
Scorecard (this week vs last week):
- Retention (rolling 30 days): [X]% vs [X]%
- Median time to first win (new clients): [X] days vs [X]
- Refunds / chargebacks: [X] / [X]
- Renewals / ascensions: [X] / [X]
- Completion, milestone [X]: [X]%
- Proof captured: [X] testimonials, [X] case studies
- Referrals: [X] leads
- Median response time to at-risk: [X] hours
- NPS (latest responses): [X]
Red clients and the plan:
- [Name]: [reason]. Plan: [action, date].
- [Name]: [reason]. Plan: [action, date].
Wins: [one or two client results worth sharing]
Renewal and ascension opportunities: [names, why they are ready]
Pattern I'm seeing: [one observation about where clients get stuck]
What I need from you: [one specific ask, or "nothing this week"]
A few rules that make this report land:
- Lead with the truth. If refunds went up, say it first, with the reason and the plan.
- Always compare to last week. A number without a trend means nothing.
- Name names for red clients. "Some clients are struggling" is not a report.
- One ask, not five. The founder acts on one.
When I read a CSM's weekly report, the "pattern I'm seeing" line is what tells me whether I have an average CSM or an S-Tier one.
When numbers move: a quick diagnosis table
| What you see | Likely cause | First move |
|---|---|---|
| Time to first win creeping up | Onboarding too slow or week one too heavy | Shorten setup, pick one easier first win |
| Refunds concentrated in month one | Expectations set in sales don't match delivery | Share refund reasons with the founder and sales |
| Attendance dropping across the book | Calls feel optional or irrelevant | Ask three clients why, adjust the call format |
| Completion stalls at one milestone | That step is too hard or unclear | Add a template, a walkthrough or a check-in at that step |
| Many greens, few renewals | Nobody is having the conversation | Build a renewal list and schedule the talks |
| Lots of wins, little proof | No proof ask in the process | Add a testimonial ask right after every win |
| Response time slipping | Book too big or process too manual | Batch check-ins, then discuss book size |
Common mistakes with client success metrics
- Tracking too much. Twenty metrics means nobody reads any of them. Start with the scorecard.
- Copying SaaS metrics. Logins and feature adoption say little about whether a coaching client gets a result.
- Changing definitions mid-stream. Your trend breaks and nobody trusts the numbers.
- Hiding bad numbers. Founders find out anyway, and then they trust nothing in your report.
- Measuring without acting. A red client with no call booked is a refund in waiting.
- Inventing benchmarks. Compare yourself to your own last month, not to a number someone posted online.
If you want to show a coach you can work like this before you are hired, a sample weekly report built on a made-up book is a strong addition to your application. The CSM Council certification also covers KPIs and founder reporting, with a record a coach can verify.
Next step
Build the sheet this week: one tab for clients, one for the weekly scorecard. Fill it for your current book, score every client green, yellow or red, and send your first weekly report using the template above. Then pick the one leading indicator that looks weakest and work on it for the next 30 days.
If you want to go deeper on the number most founders care about first, read the client retention guide next.
Frequently asked questions
What are the most important client success KPIs?
In a coaching business or agency, the core client success KPIs are retention across the book, time to first win, refund and chargeback rate, renewal and ascension rate, program completion, case studies captured and referrals generated. Response time to at-risk clients and NPS round out the picture.
How do you measure customer success in a coaching business?
Measure whether clients get the result they paid for and stay. In practice that means tracking how fast clients reach a first win, how many complete the program, how many refund, and how many renew, upgrade or refer. Track it per client in one sheet and review it weekly.
What is a customer health score?
A customer health score is a single number per client that combines a few signals, like attendance, check-in replies, progress against milestones and sentiment. Each input gets a weight, the total maps to green, yellow or red, and red clients get a call before they ask for a refund.
How do you calculate client retention rate?
Take the clients active at the start of the period, subtract those who left during it, and divide by the clients active at the start. New clients who joined during the period are left out, so new sales do not hide churn.
What is the difference between leading and lagging indicators in client success?
Lagging indicators report what already happened, like retention, refunds and renewals. Leading indicators predict them, like attendance, reply rates, time to first win and health scores. You manage the leading ones week to week and report the lagging ones to the founder.
Is NPS useful for coaching programs?
It is useful as a trend and a conversation starter, not as a verdict. Ask it at fixed points, like day 30 and at the end of the program, and always read the comments. A low score is a reason to call, and a high score is a reason to ask for a testimonial or referral.
What should a client success manager report to the founder each week?
One short page: the scorecard numbers with last week's comparison, the number of clients in green, yellow and red, the top at-risk clients with the plan for each, wins and proof captured, renewal opportunities, and one thing you need from the founder.
Written by
Diego Miescher
Founder of CSM Council and Scale With Fulfillment. Five-plus years running client success and fulfillment for coaching businesses doing $500K to $1M+ a month, and the person who hired and trained their CSM teams.