Our thinking

The questions behind trustworthy screening.

Evidence, assumptions and the questions worth asking.

A signal becomes useful when its source can be traced, its limitations are visible and the next action is clear. Our thinking explores the evidence and emerging risks shaping that standard.

Good questions make room for better evidence.

What we believe

A useful explanation does more than identify a mismatch. It shows where the information came from, what it supports and what could change the interpretation.

That is the standard we want to explore with people across hiring and screening. Which signals deserve attention? What makes corroboration independent? What should happen when a source cannot yet be verified?

How we approach the questions

We start with a specific question, examine the evidence and make the limits clear. Where an explanation remains uncertain, we keep the question open rather than turn an assumption into a finding.

Our immediate focus is the source: whether the employers, references and documents behind a claim are real, how to make what we find useful, and how to keep review fair to the people affected.

Evidence before interpretation

A signal is useful when it can be explained.

Show the source. Explain the uncertainty. Make the next step clear. A signal is a reason to look closer, not a verdict on a person.

Questions we are exploring

Clarity is the starting point. Confidence follows when the evidence can be understood and challenged.

Is screening quality a control worth measuring?

Insurers treat "we screen our people" as a control, yet one organisation's yes cannot be compared with another's. Our first piece asks whether that declared control can become a measured one.

Read the article

When does agreement become corroboration?

A website, reference and document may repeat the same account. What makes their agreement independent evidence, and how can a reviewer establish where each source began?

What makes a risk signal useful?

Relevance, context and a clear next step matter as much as detection. We are exploring how explanations and confidence limits can help people review evidence fairly.

How do we know a method is working?

Useful testing examines missed signals as well as unnecessary referrals. Evidence quality, reviewer understanding and the effect on candidates all belong in that assessment.

If you work in screening, hiring, fraud or risk and want to argue with any of this, we would like to hear it: