When your best person resigns.

Do three things, in this order. Find out the real reason before you respond to the stated one. Protect what leaves with them — the relationships, the knowledge, the half-finished work nobody else can see. Then talk to the people you would least like to lose next, before the news reaches them sideways. Almost everything that goes badly wrong after a good resignation goes wrong because the owner skipped straight to recruitment, or straight to a counter-offer, and did none of the three.

The instinct is to treat a resignation as a problem to be solved. It is more useful to treat it as information. Somebody who was good enough that you are worried about losing them has just made a considered decision about your business, and they will tell you why if you ask properly in the next two days. In six weeks, when they are working out their notice and have already mentally left, they will not.

What should you do in the first 48 hours?

Keep the first conversation short and do not negotiate in it. Thank them, say you are sorry to hear it, and ask for a proper conversation tomorrow. That gap does two things: it stops you saying something you cannot take back, and it lets them prepare an honest answer rather than a defensive one.

Then, in the second conversation, ask the question that actually produces information: when did you first start thinking about leaving? Almost nobody decides to leave on the day they apply for something else. The answer is usually a specific moment several months earlier — a promotion that went elsewhere, a project that got taken off them, a conversation with you that landed badly. That moment is the thing worth knowing. The job advert they answered last week is just the mechanism.

Two further questions are worth asking plainly: what would have had to be different, and what should I be doing for the people who are staying? Both are easier to answer honestly by someone who has already decided to go, which is exactly why this window is valuable.

Should you make a counter-offer?

Usually not, and the reason is not the money.

Money is the reason people give because it is the only one that can be said without insulting anybody. The reasons underneath it are more often the job itself, the manager, the absence of a visible next step, or the slow accumulation of small frustrations nobody ever raised. Pay more for the same job with the same manager and the same missing next step, and you have bought a delay rather than a decision.

There is a second cost that is easy to miss. A counter-offer teaches the rest of your team how pay rises are obtained in your business, and the lesson is: get an offer somewhere else. That is a very expensive thing to teach.

The exception is narrow but real. If you can name something specific that has genuinely changed — a role that has opened, a promotion you had already decided on, a pay correction you know has been overdue for a year — then say that, and say it as a correction rather than a rescue. A number invented on the day is not that. If you find yourself reaching for one, the honest reading is that you knew they were underpaid and did nothing until it cost you something. Pay and promotion covers how to set that structure so the question does not arrive this way.

What does replacing them actually cost?

Owners consistently under-count this, because the recruitment fee is the only part with an invoice attached. Here is an illustrative build-up on a £38,000 role — round numbers, substitute your own.

• Recruitment. An agency at 18% of salary is £6,840. Advertising and doing it yourself is cheaper in cash and more expensive in your time.
• The vacancy itself. Three months is a realistic gap between resignation and a new starter being in the building. That period is covered by overtime, by a temp, or by work not getting done — and the third option has a cost even though it never appears anywhere.
• Ramp-up. A new person reaching full output over three months is roughly £4,750 of salary paid against partial contribution. That is normal and unavoidable; it is not free.
• Your time and your managers' time. Thirty hours across shortlisting, interviewing, offer and induction is conservative. At a notional £45 an hour of owner time, £1,350.

That is over £13,000 for one mid-level replacement, and it excludes the parts that resist arithmetic entirely: the customer who liked dealing with that particular person, the process only they understood, the six months of context that walks out with them.

The point of the number is not to frighten you into a counter-offer. It is to change what you are willing to spend attention on beforehand. £13,000 is far more than the cost of a proper monthly one-to-one, a clear progression conversation once a year, or fixing the thing three people have now mentioned. Retention is cheap in advance and expensive in arrears, which is precisely why it loses to whatever is urgent.

What do you protect during the notice period?

Assume from day one that the knowledge in their head is the asset, not their remaining weeks of output. Treat the notice period as a handover project with a named owner and a written list, not as a countdown.

• Relationships. Which customers and suppliers deal with them personally? Introduce their successor — or you — while they are still there and still willing to make the introduction warm.
• Undocumented process. Ask them directly: what do you do that nobody else knows how to do? Then have them write it down, badly and quickly, rather than beautifully and never.
• Work in flight. Anything half-finished needs a name against it before their last week, not during it.
• The exit conversation. Hold it near the end, when the emotion has settled and the honesty is highest. Ask what you should change. Write down the answer even if you disagree with it.

Leave on good terms deliberately. Small business is small, people come back, and the way you handle someone's departure is watched closely by everyone who is staying.

What do you tell the team, and when?

Early, factually, and from you. The gap between a resignation and an announcement is filled by speculation, and speculation is always worse than the truth. Say who is leaving, roughly when, what is happening to their work in the meantime, and that you are recruiting. You do not need to explain their reasons, and you should not invent a comfortable version of them.

Then do the part most owners skip: a short, individual conversation with each of the people you would least like to lose next. Not a retention pitch. One honest question — what would make you think about leaving? — and then silence long enough for a real answer. What comes back is usually specific, cheap and entirely fixable, and the fact that nobody had asked is often the whole story. This is ordinary leadership coaching ground, and it is the highest-return hour in the fortnight.

What is the resignation telling you about the business?

One departure is an event. A pattern is a diagnosis, and the pattern is what you are looking for.

Ask whether the reason given points at something structural. If they left because there was no next step, your business has a progression problem and the next capable person will hit the same ceiling. If they left because of a manager, you have a management problem you have probably been aware of for a while. If they left because the work had quietly become something they did not sign up for, you have a role-design problem that is likely repeating elsewhere.

And ask an uncomfortable one about yourself. If losing one person genuinely threatens the business, the resignation has not created that risk — it has revealed it. That is a concentration problem, and it is the same problem whether the person leaves, falls ill or simply takes a holiday. Stop being the bottleneck and building a high-performing team both start from there, and it is a common reason owners come to business coaching in the first place.

What to do this week

• Book the second conversation, and ask when they first started thinking about leaving.
• Write the handover list before the last week, and put a name against every item on it.
• Tell the team the facts yourself, before the news travels sideways.
• Have one honest conversation with each person you would least like to lose next.
• Write down the reason they actually gave, and check it against the last two people who left. If it is the same reason, that is your next project.