Lesson 6 of 7 in Run and Read a Diagnosis
Why this matters
Unknown is the most misread state in the product, and misreading it is expensive. Treat Unknown as failure and you fund fixes for problems that may not exist, or tell a client their marketing is broken when the research simply could not see it. The app says it plainly: unknown does not mean absent or weak. Reading Unknown correctly keeps the Prioritize step honest, because priorities rest on evidence rather than on gaps in coverage.
When to use it
Every time an area reads Unknown, which will be often, especially for businesses with a small public footprint or sites that block automated access.
Before you start
- A finished run open on the Diagnostic page.
- Enough knowledge of the business's growth model to say which channels actually carry it. That decides which Unknowns are worth checking by hand.
Do this
- 1Read "Unknown" literally: this run could not verify enough public evidence for the area. It does not mean the system looked and found nothing.
- 2Read "What we could not find" in the method section. It separates areas public research could not gather enough data on from areas assessed from intake answers (such as "Repeat Customers" or "Email & Text Marketing") that have not been measured yet.
- 3Check the warnings above the report. A site that blocks the scanner, a timed-out page speed check, or a failed social scan each leaves its areas Unknown for this run.
- 4Open "Review the full evidence" and read the detail. A directory marked "Could not verify" is a failed check, which is different from "Not found".
- 5Never convert Unknown into a number. There is no overall score, no maturity index, and no hidden grade to fill in.
- 6Decide what to verify another way. Areas that carry the growth model deserve a manual check, or a connected account in Integrations before the next run; the rest can stay Unknown without harm.
What you should see
The same Unknown labels as before, each now with a reason you can name: blocked, timed out, not public, or intake-driven and not yet measured. When a run could observe nothing at all, the page says the areas remain unknown and that unknown does not mean absent or weak.
How to judge it
You can say "we do not have evidence for this area, because..." without adding "which probably means it is weak", and you have a short list of the Unknowns you will verify and how. Why is a diagnosis area Unknown? lists the common causes.
What it does not mean
Unknown does not mean zero, broken, neglected, or behind competitors. Scanner coverage is not a business grade, and the absence of evidence is not evidence of absence. An Unknown area is also not automatically a priority: whether it matters depends on the growth model, and that judgment is yours.
Try it in Signal & Science
Open the Diagnostic page and decide which Unknown areas matter enough to the growth model to verify manually.
Open in Signal & Science