Why MarginOS shows “N/A” instead of a made-up number

When MarginOS cannot compute a number honestly, it shows N/A — not a default, not an industry average, and above all not zero. This is a deliberate design position, and it is the reason a MarginOS dashboard sometimes has gaps where other tools show a confident figure. Those gaps are the product working correctly.

The position, stated plainly

A number on a dashboard is an instruction to act. If MarginOS fills a gap with something plausible, you cannot tell the difference between “we measured this” and “we guessed this” — and you will spend real money against the guess. So MarginOS reports what it can evidence, and reports the rest as unknown.

There is no fallback constant anywhere in the money math. No assumed CAC, no assumed margin, no industry benchmark quietly substituted when your data is thin.

Why zero is the dangerous answer

Zero is the tempting default because it keeps the layout tidy and the arithmetic working. It is also the one value guaranteed to mislead, because zero is a result and readers treat it as one.

• A channel with unknown acquisition cost displayed as $0 CAC reads as free customers. It is the single most expensive misreading available in this category of tool.
• A profit ratio shown as 0.0× reads as proven break-even, and argues against testing a young channel that simply has no cohort yet.
• A data-health score shown as 0% reads as a catastrophe to escalate, when the truth may be that nothing has been measured yet.
• And 100% is just as bad in the other direction — a section of your catalogue with no revenue is unmeasured, not perfectly healthy.

Every one of these turns “we don’t know” into “we know, and the answer is bad” (or worse, “and the answer is fine”).

Measured zero and unknown are different facts

This is the distinction that makes the position workable rather than merely cautious, and it cuts both ways.

If a group of customers genuinely generated no profit — or lost money — MarginOS shows that. A real zero is evidence, it is often the most important number on the screen, and it is never softened into N/A to spare your feelings. A channel reporting a profit ratio below 1 is telling you something true.

N/A is reserved for the other case: there is no basis to compute the figure at all. No customers in the cohort yet. No acquisition cost to divide by. No revenue in the denominator. The rule is simply that a missing input produces a missing answer, and a measured bad result produces a bad number. Collapsing those two into one display value is what most tools do, and it is what makes their dashboards unreliable at exactly the moments that matter.

Where you will see N/A

Unknown costs are quarantined, not dropped

The same principle governs costs that arrive but cannot be matched to anything — an invoice line for a product code MarginOS does not recognise, for example. Discarding it would silently overstate your profit; guessing where it belongs would silently misstate it per product.

Instead the cost is held in a visible quarantine with the reason it could not be placed, so your totals stay honest and the unresolved item stays actionable. See why a cost landed in Unmapped Costs.

What this costs you, honestly

This position has a real price and it is worth naming. A new MarginOS account has more gaps than a tool that fills them in. Some comparisons you want on day one are unavailable until enough history exists. If you are evaluating tools side by side, MarginOS will look less complete.

What you get in exchange is that every number you do see is one you can act on without checking it first. We think that trade is correct for decisions that move budget, and we would rather show you an empty cell than a confident wrong answer.

Common questions

Why does a competitor’s dashboard show a number where MarginOS shows N/A?

Usually because it applied a default when the real value was unavailable — a standard margin assumption, a blended cost estimate, or a zero. The number is not more accurate; it is more filled in. The question worth asking of any tool is what it does when it does not know, because that is the case where the answer changes your decisions most.

Can I set my own default so I see a number instead?

For costs, yes, and that is by design: you can supply a store-wide default cost or an override for a specific product, and MarginOS will use it while marking the result as less exact than a measured cost. That is your assumption, recorded as yours. What MarginOS will not do is invent one on your behalf without telling you.

Is N/A the same as zero in exports and totals?

No. An unknown value does not contribute to an average or a total as a zero would, because doing so would drag the aggregate toward a number nobody measured. It is excluded, and the coverage indicators tell you how much was excluded.

A ratio shows a value below 1. Is that a data problem?

No — that is a measurement, and it is telling you the cohort did not earn back its acquisition cost in the window. Data problems appear as N/A. If you see a real number, it came from real orders.

Why is my acquisition cost N/A when I definitely spent money?

Because the spend has not yet produced new customers that MarginOS can attribute to it. Spend with no attributable customers does not produce a low cost per customer — it produces no cost per customer, and the spend itself remains visible in the channel’s own columns so the waste is not hidden.

Related

• The Defensible Profit Model — how MarginOS grades cost evidence as measured, declared or modelled.
• Data Trust levels — the per-product reliability grade behind a margin.
• How MarginOS resolves blended CAC — the metric where this position bites most often.
• Why a cost landed in Unmapped Costs — the quarantine that keeps unresolved costs visible.