unit-economics

What's actually normal? The financial benchmarks that matter (and the ones that lie)

Five numbers to grade yourself on, the published band for each, and the line I watch instead.

By Samer Azar, Fractional CFO · 2026-08-01 · 9 min read

Key Takeaway: You cannot tell a healthy number from an early emergency without a reference point, and most founders were never handed one. Here are five numbers worth grading yourself on, the published version of healthy for each, and the line I watch instead. The published one will mislead you about as often as it helps. Grade yourself by Monday.

Dear reader,

Last week you drew one line under your ads. There are five more under your business.

You are looking at one of your own numbers right now. Your margin, or the months of cash in the account, or the number of days your stock sits before it sells.

If you are honest, you cannot say whether that number is healthy or the first sign of something going wrong.

Nobody ever handed you the reference point, so the gap is not your fault.

I have sat inside enough companies' books to know what normal looks like. I have also learned that the published version of normal misleads you about as often as it helps.

One click sharpens who I write these for.

Here is the honest problem with benchmarks. A number on its own means nothing. The same number next to the right reference point tells you where you stand.

The trouble is that the reference point everyone quotes is built from the wrong companies, so it points you the wrong way.

Five numbers are worth grading yourself on. For each one, the published band, then the line I watch instead.

The five, side by side: the published band, and the number to watch. Sources on the card.

How many months of cash is safe

The published answer is three to six months of runway. It sounds prudent, and it is close to useless.

Two problems with it. First, the median small business runs on far less. The JPMorgan Chase Institute studied hundreds of thousands of them and found the middle of the pack holds around twenty-seven days of buffer, less than a month.

The advice describes a world most founders do not live in.

Second, and this is the one that matters, a fixed number of months is the wrong target. What predicts trouble is the DIRECTION, not the level.

The line I watch is your worst month, not your average one. A plan that ends December down a little can trough far deeper in August, and the founders who raise against the year-end figure run out mid-year.

Build the forecast to the deepest month. Then add a line for the surprise you have not thought of yet, because the tax bill and the legal fee always come.

Watching that drift week to week, while it is still small, is the whole idea behind Cash Actions.

Is your margin healthy

The published band for a direct-to-consumer brand is a gross margin around fifty percent. Finaloop put the median near fifty-one across hundreds of seven and eight figure brands, and the public-company data lines up.

Here is where it lies to you. Gross margin is the number you quote with pride, and it hides the number that pays you.

A brand showing a forty-three percent gross margin can be running a twenty-three percent contribution margin once you load in the real cost of winning each sale: the ad spend, the influencer units, the fulfillment, the platform fees.

Gross margin counts the product. Contribution margin counts the business.

Your best channel is often your worst.

Direct looks like the highest margin because its per-unit cost is lowest, right up until you add the ads and the people and the software it takes to run it. Loaded fully, the channel you are proudest of can be the one losing money.

One more that hides in plain sight. Payment processing takes seven or eight percent off the top, and most founders never put a number on it.

If you work in finance and you know a founder who quotes gross margin and never looks at contribution, send them this issue.

It is the one distinction that changes how they run.

Are you acquiring customers profitably

Everyone repeats the same benchmark here. Keep your lifetime value to acquisition cost above three to one.

It is a fine rule, borrowed from software, and it misses the thing that kills product businesses. The ratio can look beautiful while the timing buries you.

The number I watch instead is payback. How many months until a new customer pays back what it cost to win them.

A gorgeous three-to-one ratio with a twelve-month payback is still a problem. It ties up cash you do not have for a full year before any of it comes back.

Under six months is healthy. Past a year, the ratio is lying to you.

And your acquisition cost is fiction unless it is fully loaded. The real cost includes the salaries, the tools, the agency, the samples, everything you spend to make the sale, not the ad account alone.

Are you holding too much stock

This is where the published number is most dangerous, because most sources average in grocery and discount chains that turn their shelves over constantly.

Strip those out, and by the ReadyRatios and NYU Stern data a normal consumer or specialty brand holds somewhere around sixty to a hundred days of inventory. Fast fashion runs leaner, luxury and seasonal run heavier.

Even that range is a comparison, not your target. Watch your own trend instead. The alarm is not a high number, it is the number doubling, because that is cash quietly walking out of the account and onto a shelf.

Two tells that your stock is running you rather than the other way around.

You describe your target as a round number, three months of stock, instead of a figure that comes from how fast each product sells. And you have no reorder rule, so what you hold is a result of habit rather than a decision.

One warning while you are in there. Expired and written-off stock blends into your cost of goods and looks exactly like a margin problem. Ask for it separately, or your margin analysis is built on sand.

Can you cover what you owe

The standard rule, the kind a bank like BDC will quote you, puts a healthy current ratio between 1.2 and 2.0, and below 1.0 means you owe more in the near term than you hold.

That part is useful, because it turns the whole balance sheet into one number a founder can feel. Below one, for every dollar you owe this year you hold less than a dollar to cover it.

The correction is the same as the buffer. Direction beats level. A ratio drifting down three months running tells you more than any single reading.

And watch for the version of this that fools people. A report that says cash improved this month is often just payments held back, the bill delayed rather than paid. The cash line looks better while you still owe every cent.

The current ratio sitting beside it tells the truth. I have watched a founder relax at the exact moment they should have worried, because they read the cash number and skipped the ratio next to it.

Why "normal" has to be your normal

Notice how far these move by business. In the NYU Stern data, a healthy net margin for an apparel brand sits near four percent. For a household-products business it is closer to twelve.

The same word, healthy, three times apart. A blended "average business" benchmark is worse than no benchmark, because it hands you false confidence.

So do this before you grade yourself. Find the two or three numbers that decide your model, and find the reference point for businesses shaped like yours, not the whole economy averaged into mush.

Those three working-capital numbers, your stock days, your collection days, and how long you take to pay, roll into one idea worth knowing.

It is the number of days your cash is out in the world before it comes back. Shorten it and you need less cash to run the same business.

The catch that ties all five together

Every one of these is a snapshot. You check it once, and it has moved by the time you look up. A benchmark tells you where you stood, and standing still is not the game.

A while back I wrote about a report that now sends itself to me every Monday. Point that same machine at the five numbers above, and it watches each one against its threshold, then tells you which one drifted and what to do about it.

The watch itself is boring, on purpose. Each number gets two things: a line it should not cross, and a direction it should not drift.

Every week the machine reads the number, sets it beside last week and beside the line, and stays quiet unless one of them breaks.

There is no dashboard to interpret and no wall of highlighted figures. One message a week, and only when something moved.

That is what I am building with Cash Actions, in the open, for founders who run on Shopify. It opens with a founding group of twenty, and the deposit is refundable if I do not deliver, so the risk stays mine.

The seven levers that move every one of these numbers, I broke down in full here if you want the whole map.

Do this before Monday

Write down your five numbers this weekend, before the week swallows the thought. Put the published band next to each one. The gaps are your shortlist, and the biggest gap is where to start.

Don't be a knocker-upper

Checking your numbers by hand, once a quarter, against a benchmark you half remember, is the knocker-upper way of running a business.

Before alarm clocks existed, there was a person whose job was to walk through town with a long stick, tapping on windows to wake people up for work. That job disappeared overnight when alarm clocks became cheap.

Spreadsheets are the long stick. A number that watches itself is the alarm clock.

Don't be a knocker-upper.

Take care of your cash, and it takes care of almost everything else.

Samer

P.S. Which of the five do you not know for your own business right now? Reply and tell me.

That is the one to fix first, and it is the kind of thing Cash Actions is built to keep in front of you.

Next week, the single number that ties your stock, your unpaid invoices, and your own bills into one measure of how long your cash is stuck out in the world, and the fastest way to shrink it.

Know a founder flying blind on these five numbers? Forward them this issue. The reference points alone are worth their ten minutes.

Not subscribed yet? Join the CFO Lab for the weekly build.