The metric the rest of the stack never built.
Marketing has CAC. SaaS has MRR and NRR. Customer success has health scores. Founders selling transformation have been operating without a metric for the moment that actually matters - the messy middle between purchase and outcome, where retention is decided. The Client Flow Score is the first leading indicator built for it. Three pillars produce one number that maps to one of five states - and tells you whether a client is in flow or about to disappear, weeks before it shows up in the cancellation column.
I. The case
Engagement metrics came from SaaS - logins, completions, time-on-platform. They measure activity inside a software product. They were never built for human transformation. A client can be highly active inside a portal and deeply stuck in their own work.
Satisfaction surveys came from customer service. NPS and CSAT measure how someone feels about an interaction. They were never built for whether the underlying outcome is forming. A client can rate you 9/10 and be drifting. Another can give critical feedback and be in full flow.
Renewal data is the lagging indicator. By the time it shows up, the decision was already made weeks earlier. The full diagnosis of why is in the manifesto. The short version: in the post-AI era, drift forms invisibly and stays invisible until the cancellation lands.
None of these metrics tell you whether the client in front of you is in flow or about to disappear.
That's the gap the Client Flow Score reads. Not engagement. Not satisfaction. Whether the client is actually in motion - and whether the motion is the right kind.
II. What the score captures
The score has three pillars. Each one reads a different layer of whether a client is actually moving - and each one is anchored in established behavioral science, not invented overnight.
Are you doing the work?
The work is being done. Output is shipping. Concrete actions are happening - not vague intentions, not "thinking about it." Execution captures whether the client is in motion or has gone quiet between sessions.
Implementation Intentions (Gollwitzer). The gap between knowing and doing is where most programs fail. Execution catches it from the action side.
Is it the right work?
Action without direction is busy-ness. Direction captures whether the client is pointed at what actually matters for their specific situation, and whether they sense the work is working. A client can be high-execution and low-direction - and that pattern almost always precedes burnout.
Expectancy-Value Theory (Wigfield & Eccles) + Sense of Coherence (Antonovsky). People persist when their experience feels comprehensible, manageable, and meaningful AND when perceived benefit exceeds perceived cost.
Do you believe in yourself and the process?
The early warning system. Belief is the felt sense of trust - in the client's own ability to follow through, and in the direction they're heading. When belief drops while execution stays high, burnout is already forming. By the time belief shows up in behavior, it's usually too late.
Self-Efficacy (Bandura). The single strongest predictor of behavior change in any guided program. When self-efficacy drops, no amount of content or support compensates.
When all three pillars are high, the client is in flow. The relationships between the pillars - which one drops first, which one stays high, how they move together - is what makes the Score a leading indicator instead of a snapshot.
III. What you'll see
Each state is a recognizable behavioral signature you can spot in a check-in, in a session, or in the rhythm of the work between sessions. Once you have language for them, you start seeing them in time to do something about it.
All three pillars high.
Doing the work. It's the right work. They believe. This is the target state. Renewals start forming on their own and referrals arrive without being asked for, because the program is doing what you built it to do.
High execution, low direction.
Working hard on the wrong things. Busy is not the same as moving. The client looks engaged on every metric except the one that matters - they're not getting closer to the actual outcome. Needs a strategic realignment, not more accountability.
Low execution, high belief.
They believe in the process. They are not showing up for it. Life is interfering, or they're stuck on something specific they haven't named. Catch this in week three or four and it becomes a small course-correction. Catch it in week eight and it becomes a non-renewal.
High execution, dropping belief.
The most dangerous pattern. They're doing everything right on paper. Internally they're running out of fuel. Experienced operators crash here - they push through until belief collapses, then leave warm and apologetic. By the time their behavior shifts, the exit has already been decided. The whole reason a metric like this needs to exist.
Belief below the floor.
Regardless of other scores, this client needs human intervention now. Not a check-in template. Not a nudge from a sequence. A real conversation with the founder. Crisis is rare, but when it happens, no automated layer is the right answer.
The whole reason a metric like this needs to exist is to catch Burnout Risk while it can still be reversed.
IV. Timing
The patterns the Client Flow Score is built to surface aren't new. The three pillars existed before AI. The five states have always been recognizable to founders who paid attention.
What changed is that drift is now invisible until it's already a decision. AI rewired how clients move through programs - they bypass content, arrive pre-educated, produce more output than founders can validate, and process resistance before they ever bring it to a session. By the time conventional engagement metrics flag a problem, the client has already mentally exited.
What's needed is a leading indicator that reads what's actually happening for the client, not what looks fine on the surface. Marketing already has it. SaaS already has it. Founders selling transformation have been operating without one.
The first transformation business to standardize a momentum metric wins the decade.
V. Where to start
You don't need the full Client Flow Score system to start measuring what it surfaces. You need three weekly questions and the discipline to track the answers per client over three to four cycles.
"How is your momentum feeling right now?"
"How much do you trust yourself and the process right now?"
"Are you doing the things you said you would do?"
Track these three answers weekly per client. Patterns reveal themselves in three to four cycles. The gap between question one and question three is often where the most useful information lives. A client who rates implementation 3/10 while writing "great week!" is giving you the real story underneath the polite one.
You're not yet measuring the Client Flow Score. You're noticing what the Score is built to surface.
VI. What measuring CFS unlocks
VII. What's coming
I built FlowOS Pulse because the manual version of this metric works - but asking founders to keep up with it week after week is the same problem we're trying to solve. Pulse runs the measurement so you read the signal, not the input form. Same three questions, automated weekly, surfaced at the moment the answer changes.
It's running with a small set of clients while the methodology matures. Until it ships more widely, every founder tracking the three questions manually is already inside the methodology - they're just doing the version of the work Pulse will take over.
When that timing changes, you'll be the first to know. The weekly letters at clientflow.substack.com are where the build is being documented in public.
The Client Flow Score is the operational layer. Three places to go next:
Reference
The vocabulary of the Client Flow Score, in one place.
Reference
Common questions about the Client Flow Score - what it is, what it isn't, and how to start using its vocabulary right now.
The Client Flow Score is the first leading indicator of client momentum built specifically for online programs selling transformation - coaching, consulting, courses, memberships, group programs. Rather than measuring activity (logins, completions) or satisfaction (surveys, NPS), it captures whether a client is actually in flow: doing the work, doing the right work, and believing in the process. It surfaces drift weeks before it becomes a non-renewal.
Every other category built a metric for the moment that actually matters in their business. Founders selling transformation never did. The gap between purchase and outcome - the messy middle where retention is actually decided - has been measured with borrowed tools. Engagement metrics from SaaS measure software adoption, not human transformation. Satisfaction surveys measure politeness, not progress. The Client Flow Score is the metric built for that specific gap.
NPS and satisfaction surveys measure how a client feels about you and the program. The Client Flow Score measures how a client is actually moving through their own transformation. A client can rate you 9/10 and still be drifting. A client can give a critical review and still be in flow. The Score isolates momentum from sentiment.
Execution (are you doing the work?), Direction (is it the right work?), and Belief (do you believe in yourself and the process?). Each pillar comes from established behavioral science - Gollwitzer's implementation intentions, Antonovsky's sense of coherence, Bandura's self-efficacy. The three together describe whether a client is actually in flow, or only appearing to be.
Flowing (all three high - the target state), Grinding (high execution, low direction - working hard on the wrong thing), Drifting (low execution, high belief - life is interfering), Burnout Risk (high execution, dropping belief - the most dangerous pattern), and Crisis (belief below the floor - needs human intervention now). Each state has a recognizable behavioral signature you can spot before it shows up in cancellation data.
Because by every conventional metric, a Burnout Risk client looks like your best client - high output, consistent attendance, warm replies, the works. The only thing dropping is the thing your dashboard is not tracking: their internal sense of trust in the process. Experienced operators crash here. They push through until belief collapses, then leave warm and apologetic. The Client Flow Score exists primarily to catch this pattern before it does.
Start by asking three questions in your weekly check-ins: "How is your momentum feeling right now?" (1-10), "How much do you trust yourself and the process right now?" (1-10), and "Are you doing the things you said you would do?" (1-10). Track them weekly per client. Patterns reveal themselves in three to four cycles. You are not yet calculating a Client Flow Score - you are noticing what the Score is built to surface.
The named components, the five states, and the scientific anchors are public - this page documents them in full. The weighting and the calibration that makes the Score actually reliable is proprietary IP, used inside FlowOS. The point of publishing the framework is to give founders a working vocabulary for what conventional metrics miss, not to ship a DIY scoring sheet.
FlowOS Pulse is the system that automates weekly Client Flow Score measurement and surfaces the right signal at the right moment for the founder. It is currently running with a small set of clients and will become more broadly available as the methodology matures. Founders who are tracking the three questions manually now will be ready for the automated version when it ships.
Founders, coaches, consultants, course creators, membership owners, and group-program operators - anyone whose business depends on clients actually completing a transformation, not just buying access. Programs where the client's transformation is the product, and where a client who stays is worth more than a new one. Low-value programs delivering one-off packages don't need this metric.
One letter per week. The Client Flow Score was developed in public, week by week, before it became this page. Subscribe and you're inside the build.