Moving On
What a reference check actually is, and what it can say
The reference stage is far more constrained and far less mysterious than candidates imagine, and knowing its mechanics removes most of the anxiety.
By Julien Perrot4 min read

Two different things called a reference
The first is a factual confirmation, usually handled by an HR function or an outsourced screening service. Dates of employment, job title, sometimes whether you would be eligible for rehire. It is a compliance step and it’s deliberately unrevealing, because many organisations have concluded that saying anything beyond the facts creates risk they would rather not carry.
The second is a conversation between two people who work in the same field, one of whom knows you. This one has content, and it is the one that actually influences a decision. It may be arranged formally, or it may be a phone call that happens because the hiring manager knows somebody who knows you, which is common in smaller professional worlds and isn’t something you can control.
What a former employer can and can’t say
Rules here vary substantially between countries and are frequently misdescribed. It is widely believed that an employer may only confirm dates, and in many places that’s a policy adopted for caution rather than a legal prohibition. Broadly, references in a good many jurisdictions must be accurate and not misleading, which is a different constraint from being forbidden to say anything negative. Some countries require a written reference of a specified form; others require none at all.
Because the position genuinely differs by place, treat any confident general claim about what an employer is allowed to say with suspicion, including one you read online. If your circumstances make this important — a disputed departure, a disciplinary matter, a dismissal — that is a moment to get advice from someone qualified in your jurisdiction rather than relying on custom.
What the conversation is actually for
The hiring manager has already largely decided, in most cases, and is doing one of two things. Confirming that nothing alarming exists, which is the usual purpose and is why most reference calls are short and cordial. Or resolving a specific doubt raised in interview: whether you can really operate independently, whether a particular achievement was as much yours as it sounded, whether the reason for leaving matches what you said.
That second purpose is why the most damaging thing in a reference is rarely criticism. It is a mismatch. If you described a role as leading something and your referee describes you as contributing to it, the discrepancy does the damage, not the referee’s opinion. Hesitation has a similar effect: a pause before answering whether they would work with you again communicates more than a sentence would, and it is not something anyone can be held to.
Choosing and preparing referees
Choose people who saw your work closely and can speak in specifics, which usually means a direct manager or a senior colleague rather than the most impressive name available. A senior person who barely knows you gives a generic answer that reads as faint, and that is worse than an enthusiastic account from someone mid-level who can describe what you actually did.
Always ask first, and ask in a way that makes refusal easy — whether they would be able to give a strong reference, rather than whether they would give one. Someone who hesitates is doing you a favour and should be thanked rather than pressed. Then brief them: the role, what it needs, the two or three things you would like them to be able to confirm, and any question likely to come up. That is not coaching them to lie. It is making sure a busy person recalls the relevant year rather than the most recent thing they remember.
Keep the relationships warm before you need them. A message once or twice a year to a former manager costs nothing and is the difference between a warm reference and one given by somebody reconstructing you from memory. This is the practical reason that leaving well matters years later.
When the last job ended badly
If the departure was difficult, the useful step is to find out what will actually be said rather than to hope. Many organisations have a written policy stating exactly what they disclose, and asking HR for it is an ordinary request. Where a settlement was involved there may be an agreed form of words, which is often the point of including one.
You are also not obliged to offer your most recent manager as a referee if the relationship was poor, though a gap in the sequence invites a question, so have a short calm explanation ready. And be honest with yourself about the risk in informal checks: they happen, they are unregulated, and no policy governs a private phone call between two people who know each other. That is not a reason for anxiety in most cases, but it is a fair reason to avoid burning a relationship on the way out even when the provocation is real.
Common questions
Can a former employer stop me getting a job?
A deliberately false or misleading reference can be legally actionable in many places, but proving it is difficult and slow. In practice the more common harm is a lukewarm or hesitant reference, which is not actionable anywhere. If you have reason to think something inaccurate is being said, take proper advice locally.
Do I have to give my current employer as a referee?
Usually not until an offer is made, and a request to delay contact until then is standard and readily accepted. Employers understand that approaching a current employer early can jeopardise your position, and an offer is normally made subject to references rather than after them.
Will they find out I have applied elsewhere?
Not through a formal reference process, which does not begin until you have consented and normally not until an offer stage. Informal checks are a different matter, particularly in small industries, which is a reason to be thoughtful about who you tell during a search.
Deputy editor, After the First Job
Julien covers first months, managing up, money at work and the questions readers actually send in and would rather show the working than assert the conclusion.





