Financial health · One number
One score,
and the five figures it is made of.
A single score compresses a business into a figure nobody can check, and the compression is where the dishonesty gets in. Two rules hold this one up: every component is already on the dashboard, and a component the product cannot measure is dropped rather than guessed.
Measured, not estimated · EU-hosted · Read-only by default
- Components5margin, runway, ads, refunds, growth
- Each oneon the screenalready
- Unmeasurabledroppednot guessed
- Under half the weightno scoreand it says why
- Net margin10031.8% of every euro sold is kept
- Cash runway100Nothing is burning: the store adds cash each month at this rate
- Ad return1002.13 in profit back for every 1 spent
- Refunds913.2% of orders came back
- Growth51Revenue up 0.4% on the period before
The score is never shown without its parts. Compression is where dishonesty hides, so everything that went into the number is on the same screen as the number.
If fewer than half the components can be measured, no score is shown at all — an incomplete score is worse than none.
01 / The case for it
A single score is where the dishonesty gets in.
Without it
Compress a business into one number and nobody can check it. Worse, a store the product can only half see gets the same confident figure as one it can see completely, and nothing on the screen distinguishes them.
What Profflow does
Five measurements you already have, and it says what it saw.
Margin, runway, ad return, refunds and growth, each scored on a band written down in advance rather than tuned until the demo looked good. Anything it cannot measure is dropped and the weights renormalised — not scored as a zero.
02 / Worth understanding
Why a score is dangerous on its own
A single figure compresses away the argument
Scores are popular because they are easy to read and hard to check. This one is only defensible because the five components, their weights and their measurements are printed beside it — so you can disagree with the score and see exactly where.
What cannot be measured is dropped, not filled
A component with no data is removed and the weights are redistributed, with the omission stated. Filling it with an average would make the score look complete and quietly make it wrong.
Under half the weight, no score
If most of the inputs are missing, the product declines to produce a number. Refusing to answer is the correct answer more often than product design usually allows.
03 / What it lets you do
What follows from getting Financial health right.
Bands written down
Not tuned until the example store looked good. The band a figure falls into is visible and the same for every store.
Components dropped, not guessed
No cash figure means runway is not scored. No ad spend means the ad return is not scored. The remaining weights are renormalised.
It says what it could see
A store scored on three of five components is not the same claim as one scored on all five, and printing the same 74 for both would be the compression doing exactly what it is accused of.
A floor under the score
Below half the total weight there is no score. At that point the honest answer is “not enough is known yet”, and a number would be worse than the sentence.
Re-derivable by hand
The score is the weighted average of the bands. Nothing else feeds it, so nothing about it is a black box.
Movement, not just level
Where the score went since the period before, and which component moved it.
04 / What it is built from
Financial health is assembled
from accounts you already run.
- 01Net margin
- 02Cash runway
- 03Profit ROAS against break-even
- 04Refund rate
- 05Growth on the period before
- 06The bands, written down
- Bands written down
- Components dropped, not guessed
- It says what it could see
- A floor under the score
05 / How it is worked out
From the account
to the figure on the screen.
- 01ScoresEach component on a band that is written down rather than tuned.
- 02WeightsThem into one average, and shows the weights.
- 03DropsAnything it cannot see, and renormalises what is left.
- 04RefusesTo score at all below half the total weight.
[ Asked of the figures ]
What is a 74 actually made of?
Five bands, weighted: net margin, cash runway, ad return, refund rate and growth. The score is their weighted average and nothing else, so it can be re-derived by hand from the same screen.
How Profflow reasons06 / How the figure is kept honest
The same four rules,
on every screen.
- Read-only, on every plan
Profflow observes the accounts it is connected to. It does not write an order, a price, a campaign or a payout back to any of them.
- Measured stays measured
A charge read from an account is labelled measured. Anything modelled — a forecast, a projection, a filled gap — is labelled as such on the screen it appears on.
- Every figure opens
Follow any number to the dated rows underneath it. A figure nobody can check is a figure nobody acts on.
- You approve
The product prepares a decision and shows the evidence. Nothing consequential happens until you say so.
About
Financial health.
Can’t find the answer to your question?
Contact us by clicking here.
No. It is a reading of five measurements you can already see, arranged so a change in any of them is visible in one place. It is not a rating and nobody outside your account sees it.
Because the figure behind it does not exist yet. Runway needs a cash balance; ad return needs ad spend. Rather than score a zero, the component is dropped and the score says so.
No, and it is deliberately not built to. There are no benchmarks in it, because a benchmark drawn from stores nothing like yours is a comparison that misleads.
It means five measurements are in good bands. The alerts, the anomalies and the cash forecast are where the specific problems live.
See Financial health on your own figures.
A single score compresses a business into a figure nobody can check, and the compression is where the dishonesty gets in. Two rules hold this one up: every component is already on the dashboard, and a component the product cannot measure is dropped rather than guessed.