Credential reuse rate, undefined, which is a different thing from zero or small

Any reuse-rate figure quoted for Solidus today would be fabricated.

What the metric is

How many times a credential is re-presented after it is issued. It is the n in the total-cost-of-ownership arithmetic, and it is the variable the entire commercial argument turns on.

• High reuse, a person verified once, presenting many times, across refresh cycles and across businesses, and the portability model is cheaper than re-running a pipeline every time.
• Reuse of zero or one, and the model is more expensive, because the buyer paid for a credential apparatus and then used it the same number of times they would have run a check.

The crossover we publish is a function of this number, and we do not have it.

Undefined is not the same as low, and the distinction is not pedantic

A low reuse rate would be a measurement, a real denominator, a real numerator, a disappointing answer. We do not have a disappointing answer. We have no answer.

There is no denominator. No production issuance volume, no relying parties presenting to, no dashboard collecting it. The metric has never been computed because there is nothing to compute it over.

Saying "our reuse rate is currently low" would sound more honest than it is, it would imply measurement that has not happened.

What would actually drive it, and how much of that is outside our control

• How many businesses accept the credential. This is the dominant term, and reliance records the count of unaffiliated relying parties that have accepted one in production: zero. With one accepting business, reuse is confined to that business's own refresh cycles. Cross-business reuse, the version of the argument that matters, requires a network that does not exist.
• The refresh cadence of the regulated activity. A re-KYC obligation that recurs generates guaranteed repetition; a one-time onboarding does not.
• Credential lifetime against that cadence. A credential that expires before the next required check is not reused; it is reissued, which is a different and more expensive event.
• Whether the holder still has it. Wallet retention over twelve months is a real attrition question and we have no data on it whatsoever.

Three of those four are decisions other people make. The metric our commercial case depends on is one we cannot measure and largely cannot control: that is an uncomfortable thing to write on our own economics page, and it is the accurate description.

Even the definition is unsettled

Before a number could mean anything, we would have to fix what it counts: presentations per issued credential? Per verified person? Per relying party? Over what window? Do re-presentations to the same business count the same as a presentation to a new one, when only the second is evidence for portability?

We have not settled that, and a rate published against an unstated definition is not a measurement, it is a choice.

What has to exist before we publish one

Production traffic; at least one relying party that is not us; a settled definition; and a dashboard that computes it from real events rather than from a spreadsheet. None of those exists today.

When they do, we will publish the number including if it is bad, and until then, the absence of a figure on this page is the finding.