Two candidates, two background checks. The first comes back clean: nothing found. The second comes back with a flag, a resolved matter from years ago that the candidate actually mentioned themselves in the interview. Most hiring managers relax about the first and start to worry about the second.
That instinct is often exactly backwards. “Clean” on the first candidate might mean there’s genuinely nothing there. Or it might mean they have a common name the search couldn’t pin down, an identity nobody really verified, or a history in a country you never checked. The flag on the second, read in context, is minor, old, resolved, and, crucially, disclosed, which tells you something reassuring about how the person handles their own past.
This is the part of background screening nobody really teaches: how to read the results. And the stakes are rising. HireRight’s 2025 Global Benchmark Report found that nearly one in five background checks turns up a discrepancy between what the candidate claimed and what the records show, and more than three-quarters of employers caught at least one discrepancy in the past year. Gartner has gone further, projecting that by 2028 as many as one in four job candidates worldwide could be fake, built from stolen identities or AI-generated profiles. More checks will come back with something on them, and reading that something correctly is the whole game.
Because a background check doesn’t hand you a verdict. It hands you evidence. The work is interpreting it: telling a genuine red flag from a reassuring green one, and both from the large middle that’s just context or noise. And the first myth to clear out of the way is the most common one, that a clean check is automatically good news.
A background check is evidence, not a verdict
The mistake underneath most screening errors is treating the report as the decision. It isn’t. A check is a set of findings, and findings have to be read, weighed, and placed in context before they mean anything. Hand the same report to a nervous manager and an experienced investigator and you’ll get two different decisions, not because the facts changed, but because reading them is a skill.
It helps to sort everything a check returns into three buckets. Red flags are signals that should genuinely worry you. Green flags are signals that should reassure you. And in between sits a large middle of information that is neither until you add context, an old record, an employment gap, a lawsuit, a news mention. Most guidance only ever talks about the red. A good read uses all three, because the goal isn’t to find something wrong, it’s to understand who you’re actually dealing with.
What’s actually a red flag
Start with the reds, but be precise about what earns the name, because “negative” and “red flag” are not the same thing. A genuine red flag tends to be relevant to the role, material rather than trivial, corroborated rather than rumoured, and, best of all, a pattern rather than a single stray incident.
The ones that reliably matter: a material discrepancy, a fabricated degree, an invented job, a title or dates stretched well beyond the truth, especially when it’s recent and clearly deliberate. Undisclosed criminal or regulatory history that’s relevant to the role. Adverse media tied to the person, not just a namesake, covering fraud, misconduct, or mismanagement. Sanctions or politically exposed exposure. A director disqualification. And one that’s easy to overlook: evasiveness itself, a candidate who resists verification, can’t explain a gap, or gets their own history wrong is telling you something. The strongest red flags rarely stand alone. One odd date is a mistake; a career’s worth of them is a pattern. And a pattern is worth more than any single finding, because it tells you about behaviour rather than a single moment.
What’s a green flag
Here’s the half almost every article on this subject skips. Screening isn’t only about catching bad actors; it’s about confirming good ones, and a mature read looks for the signals that should reassure you as deliberately as the ones that should worry you.
The green flags worth weighing: a consistent, independently verifiable track record, where what the candidate said matches what the records confirm. Transparency and proactive disclosure, because someone who volunteers a past issue is very often lower risk than someone who quietly hides a small one. Corroboration across independent sources, rather than a story that only holds together from one angle. Stable, verifiable professional associations. A clean regulatory record in a field where regulators actually keep score. And a simple willingness to be verified, which sounds trivial until you meet someone who resists it. Green flags do real work: they let you move quickly and confidently on strong candidates, and they’re half of what a background check is actually for. A programme that only ever hunts for reasons to reject is doing half its job, and the slower, more painful half at that.
Clean is not green
Which brings us to the trap that catches careful people, the quiet assumption that a clean result is a positive one. It isn’t. The absence of a red flag is not a green flag. It’s an absence, and an absence has more than one explanation.
“Nothing found” can mean exactly that: nothing is there. But it can just as easily mean you searched the wrong name, the wrong jurisdiction, or the wrong language, and missed what was there. It can mean the identity was never properly verified, which is the fastest-growing gap of all now that fake and stolen identities are a routine part of hiring fraud. Or it can mean the person is genuinely new, with no track record yet, and no data is not the same as good data. Treat a clean check as what it is, “no red flags surfaced,” rather than what it feels like, “verified safe.” The two are a long way apart, and the distance between them is where a lot of bad hires live.
Context changes the colour
Most of what a check returns isn’t a clear red or green at all. It’s raw information that turns one colour or the other only once you add context, and reading that context well is most of the job.
A criminal record is the obvious case: its meaning depends on the nature of the offence, how long ago it was, and whether it’s relevant to the role, which is why a thoughtful, individualised read beats a blanket rule every time. An employment gap is not a red flag on its own; it can be caregiving, illness, study, redundancy, or a dozen ordinary things, and treating every gap as suspicious mostly just screens out people with lives. A lawsuit means something different depending on whether your candidate was the plaintiff or the defendant, and over what. A piece of adverse media depends on its source, its stage, and whether it’s even the right person. The same raw finding can be red, green, or nothing at all depending on the facts around it, and the reflex to colour it in before you’ve read it is where good screening goes wrong. The discipline is simple to state and hard to keep: read first, colour second, never the other way round.
What you’re not allowed to treat as a red flag
Reading red flags responsibly isn’t only good judgment; in many places it’s the law, and getting it wrong creates a different risk entirely. A few guardrails matter.
You can’t base a decision on protected characteristics, race, religion, age, disability, national origin, and the rest, no matter where in a report you happened to see them. Under the standards the EEOC enforces, a criminal record generally can’t be an automatic bar; it calls for a job-related, individualised assessment that weighs the nature of the offence, the time passed, and its relevance to the role, and many jurisdictions add “ban-the-box” rules on top. When a third-party provider runs the check, the Fair Credit Reporting Act sets out disclosure, consent, and adverse-action steps you have to follow, and in the EU the GDPR requires a lawful basis and proportionality before you process someone’s data at all. The through-line is simple: decisions have to be relevant, consistent, and documented, which, usefully, is also what makes them defensible.
Reading checks well, at scale
Doing this for one candidate, carefully, is manageable. Doing it consistently across every candidate, so that the same finding is read the same way each time and the reasoning is written down, is harder, and it’s where programmes either hold up or quietly fall apart. The answer is a framework applied the same way every time: is it relevant, is it material, is it corroborated, is it recent, and is it actually the right person?
It’s also worth knowing the limits of the check in front of you. A standard background check is narrow by design, identity, criminal records, credential verification, and it has a ceiling: it mostly surfaces what resulted in a conviction or a formal record, and misses the adverse media, regulatory actions, and litigation that often signal risk earlier. For executives, board members, and other high-exposure roles, the reds and greens that matter most usually live above that ceiling, which is a case we made in full in our piece on negligent hiring. The deeper the exposure, the more the read depends on information a basic check never collected. The higher the stakes, the more the real signal, red or green, tends to live in the part of the record a standard check was never built to reach.
Read the whole picture, not just the alarms
A background check was never meant to make the decision for you. It hands you evidence, and the value is in reading it well: separating the genuine red flags from the reassuring greens, and both from the ocean of context and noise, consistently and in a way you can defend. A clean result isn’t a green flag, a disclosed old matter isn’t necessarily a red one, and most of what you’ll see is neither until you’ve done the reading.
Beady is built to surface the signals a standard check misses and to make them readable. Its adverse media screening brings together adverse media, sanctions and PEP exposure, litigation, and regulatory actions on a person and the entities around them, with each finding linked back to its original source so you can weigh it in context rather than react to a bare hit. It resolves identity, so a clean result actually means clean rather than a namesake you missed, and it covers the people behind companies, not just the name on the application. It doesn’t replace your background check; it tells you how to read the half the check can’t see.
If you’d like to see the reds, the greens, and everything in between on someone you’re weighing right now, book a demo and we’ll walk through it with you.
