When we cannot find out enough about someone

What a public profile leaves out

Before anybody is contacted, the platform has to build a picture of each candidate out of what is publicly available about them. Everything downstream leans on that picture, so the honest question is not how good it is on an average day. It is what happens on the day it comes back almost empty.

Public footprints are uneven for reasons that have very little to do with ability. Somebody who has been at the same company for nine years has less written about them than somebody who moves every eighteen months. People in some functions publish constantly and people in others never do. A candidate who keeps their profile deliberately sparse is quite often the senior one who has not needed to be findable for a decade.

One source, and no pretending it worked

Enrichment reads each candidate through a single source. There is no second purchase waiting behind it, so the question that matters is what happens when that source cannot read a profile. People search is different: there, two data providers back each other up, so a search does not fail because one of them has a gap.

Enrichment is not like that: when its one source cannot read a profile, no further lookup is bought automatically; pulling the full profile is your call. The care goes into what the platform does with the miss instead.

When the profile cannot be read

Sometimes the profile comes back empty, and this is the point at which most systems quietly go wrong. The tempting behaviour is to score whatever is on file. A candidate with almost nothing recorded scores badly against a list of requirements, because there is no evidence they meet any of them, and the result is indistinguishable from a candidate who was properly assessed and found wanting.

Those two are not the same thing, and the difference bites hardest among the people you would most want to speak to. So when the picture cannot be built, the candidate is kept as a preview profile and unscored rather than scored against nothing or silently failed, and the record says as much. That state is not a poor result. It is the absence of one.

A candidate nobody could find out about has not failed your requirements. Nothing has been measured yet.

Why you are shown it rather than spared it

A system could hide preview-profile candidates, and there is a defensible argument for doing so: you asked for people who match a brief, and this person cannot be shown to match it. It would also mean an entire category of candidate vanishes from your search without you ever learning that the category was there.

So “we could not find out” is its own state, sitting next to the candidates who carry a real assessment rather than being folded in among them. You can see how many there are. You can open one and see that the gap is in what was available to read, not in the person.

Coverage gaps are not randomly distributed

The reason this is worth the engineering rather than a note in a help page is that missing data is not scattered evenly across a market. It clusters. Particular industries, particular seniorities, particular languages and particular parts of the world are consistently thinner in public sources than others, and the thinness is a property of the source rather than of the people in it.

A pipeline that scores absence as failure therefore does not produce a slightly noisier shortlist. It produces a shortlist that leans, every time, in the same direction, towards the people who happen to be well documented. Nobody chose that, and because the effect is silent nobody would ever see it in the output.

Keeping the preview state separate is what stops that pattern accumulating quietly across every role you run. It does not make the missing data appear. It keeps the gap visible long enough for a person to decide what to do about it.

What to do with a preview-profile candidate

• Start with the name and the current role. Those two survive even when everything else is missing, and they are frequently enough to tell you whether this is worth a minute of your attention.
• Read a sparse footprint as a fact about the market, not the person. In some functions and some geographies the strongest people are simply the hardest to read from the outside, and a search that filters them out is not a more accurate search.
• Decide on purpose. Looking at a preview-profile candidate and moving on is a perfectly good outcome. Never having been shown them is not, because that decision was made for you by a lookup that happened to return nothing.