Twelve statements, and one number you already know.
Score each statement out of ten for how true it is today, across four areas — knowing what is expected, being able to do it, straight conversations, and reasons to stay. Then give the tool four figures about the people who left last year. You get the weakest area with the first thing to do about it, and what the leavers cost in recruitment and lost output on your own numbers. It updates as you type, and nothing you enter leaves your browser.
Where the team actually is.
The weakest area matters more than the score out of 120.
Three things on the panel are worth more than the total, and they are the three a coaching conversation about your team starts from.
The weakest area
The panel names one area of four and one action. A team problem broken into twelve pieces gives you nothing to do on Monday, so the tool commits to the lowest-scoring area and hands you the first move rather than a plan.
The three lowest statements
Areas average out and can hide a single bad answer. The three lowest individual statements are usually more recognisable than the area name, and they are normally the sentences you hesitated over.
The cost of the leavers
The scores are your opinion. The people who left are a fact, and the figure puts recruitment and lost output against them. It is the number that decides whether fixing the weakest area is urgent or merely sensible.
Most of this needs work. Usual in a team that grew faster than the way it is run, and not a crisis — but pick one area and give it the whole quarter rather than starting four.
A team that functions, carrying real drag. The weakest area is costing you hours every week, and it usually shows up as work you end up redoing yourself.
Strong. At this level the constraint is normally one area or one person rather than a general condition — and it is worth having two of the team score the same twelve statements.
What does the Team Health Check measure?
Four things, with three statements each. Whether people know what is expected of them. Whether they are able to do it. Whether the conversations between you and them are straight ones. And whether they have reasons to stay. Between them those four cover most of what goes wrong in a team of three to thirty, and they are in that order deliberately, because the later ones rarely hold while an earlier one is broken. Praise does not rescue somebody who cannot say what they are responsible for. No amount of one-to-ones will settle a person who has been asked to do a job they were never equipped to do.
Scoring out of ten rather than answering yes or no is the same 0–10 rule used everywhere else in the coaching, and for the same reason. “Do your people know what is expected of them?” invites a defensive yes. “Out of ten, how true is that?” produces a number you can compare against the other eleven and move by one. And one point is the right size of ambition: a real piece of work with a date on it, rather than a transformation that gets quietly abandoned.
What does the tool do with your answers?
The arithmetic is simple on purpose, because the value is in the honesty of the twelve numbers rather than in the model. It adds them for a total out of 120 and divides by twelve for your average. It averages each area of three and names the weakest, breaking ties by the order the areas appear so the same answers always produce the same result. It lists the three lowest individual statements, and counts how many of the twelve sit below your own average — which tells you whether you have one collapsed area or a general condition.
If all twelve come out at the same number, the tool says so and refuses to nominate an area. Nothing would be weaker than anything else, so naming one would be a recommendation your own answers do not support.
What does a leaver actually cost?
Two things you can put a figure on without guessing, and four you cannot. The two: what you spent to replace them, and the share of a year’s salary you paid while the replacement was still getting up to speed. Twelve weeks of ramp-up on a £32,000 salary is £7,385 of pay against work not yet at full output, and with £2,500 of recruitment cost that is £9,885 for one leaver. Two of them is £19,769, which for a team of eight is a quarter of the team in a year.
The four it leaves out are all real: your own hours interviewing, inducting and covering the gap; the overtime and goodwill the rest of the team spent absorbing the work; any client who left with the person or during the gap; and the way one resignation makes the next easier for everybody else to imagine. That is why the figure is built to be a floor rather than an estimate. A floor is the right way round for a number you might use to justify spending money, or time, on keeping the people you have. If you want the other side of the same arithmetic — what a new person has to earn before they pay for themselves — that is the true cost of a hire calculator.
A worked example
The figures the tool opens with are an invented profile rather than a client. Knowing what is expected scores 7, 4 and 3; being able to do it 6, 3 and 5; straight conversations 3, 4 and 6; reasons to stay 5, 7 and 4. That is 57 out of 120, an average of 4.8, with six of the twelve statements below that average.
Straight conversations is the weakest area at 4.3, so that is where the work starts: a half-hour one-to-one with every person, monthly, kept in the weeks you are busiest. Look at the shape though, because it is a common one. This owner scores 7 on people being able to state what they are responsible for, but 3 on anybody naming the quarter’s priorities — so individual jobs are clear and the direction is not. They score 3 on anyone covering the busiest role for a fortnight, which is why the owner cannot take a holiday. And they have lost two people out of eight in a year, at a cost of £19,769 on their own figures. An owner with this profile usually believes they have a recruitment problem. The twelve statements say they have a management-rhythm problem, and that recruiting into it will produce the same result again.
What do you do with the answer?
Write one sentence: what would make the weakest area one point higher by the end of this quarter? Put a date on it, tell one person, and leave the other three areas alone until that one has moved. Then hand the same twelve statements to two or three of the team, ask them to score independently, and talk about every statement where you are more than two points apart. Where they score lower than you, they are living with something you have stopped seeing.
Some of what comes out of this you can fix on your own: a monthly one-to-one in the diary, a line written down for each person saying what they own. Others — handing over work you have always done yourself, or the conversation with somebody whose work has not been good enough for a year — are the ones owners circle for months. That is what leadership coaching is for, and the first conversation is free. If the wider business is what is really constrained rather than the team, the Wheel of Business scores all nine areas in about four minutes, and the when to hire guide covers the decision this tool only costs.
What the check assumes
- It is a self-assessment, and it is exactly as good as the honesty of your twelve numbers.
- All twelve statements are weighted equally. In your team they are not equally important, which is why the tool names a starting area rather than issuing a plan.
- Nothing is scored against an industry benchmark, because no honest benchmark exists for a twelve-statement self-rating. Every comparison is against your own average.
- The cost of a leaver uses only the figures you type in. There is no assumed recruitment fee, no assumed salary and no assumed ramp-up period.
- Nothing you type leaves your browser. There is no database behind this page and no answer is stored.
Want the longer version of this?
We will send you the write-up: what a strong and a weak answer looks like on each of the twelve statements, and how to run the same check with your team without it turning into a complaints session.
Your answer is already on this page — the email is optional. We only use it to send the write-up and the occasional guide, and the message includes a short summary of your own result so the follow-up is about your team. Unsubscribe in one click. See our privacy notice.
Frequently asked questions
What does the Team Health Check measure?
Four things, three statements each. Whether people know what is expected of them. Whether they are able to do it. Whether the conversations between you and them are straight ones. And whether they have reasons to stay. Those four cover most of what goes wrong in a small team, and they are in that order on purpose, because the later ones rarely get fixed while an earlier one is broken. Somebody who cannot say what they are responsible for will not be rescued by praise, and no amount of one-to-ones will settle a person who has been asked to do a job they were never equipped for.
Should I score this myself or ask the team?
Do it yourself first, because it takes four minutes and it tells you what you believe. Then hand the same twelve statements to two or three of the team and ask them to score independently. The gaps are the whole point. Where they score lower than you, they are living with something you have stopped noticing. Where they score higher, you may be carrying a worry the team has already solved. Do not average the two sets of numbers together and do not go looking for who said what — take every statement where you are more than two points apart and talk about that one.
Why does it ask about people who left?
Because the score is an opinion and the leavers are a fact. Twelve honest numbers tell you where the team is weak; the people who walked out of the door last year tell you what that weakness already cost, in money you have spent rather than money you might lose. The two halves check each other. A comfortable score alongside three unwanted resignations means the scoring was generous. A poor score with nobody leaving means you have time to fix it, which is a genuinely better position than it feels like from the inside.
How is the cost of a leaver worked out?
Entirely from the four figures you type in, with no assumed rates and no benchmark. It takes the recruitment cost you give for one replacement, then adds the share of a year’s salary covered by the weeks it takes a replacement to become fully productive — twelve weeks on a £32,000 salary is £7,385 of pay against work that is not yet at full output. Those two together are the cost of one leaver, multiplied by the number who left. It is deliberately conservative arithmetic, and the page lists what it leaves out.
What does the cost figure leave out?
Four things, all of them real and none of them safely guessable. Your own hours spent interviewing, inducting and covering the gap. The overtime and goodwill the rest of the team spent absorbing the work in the meantime. Any client who left with the person or during the gap. And the slower, harder-to-see cost of the next resignation becoming easier to imagine once one person has gone. The figure the tool gives you is therefore a floor rather than an estimate, which is the right way round for a number you might use to justify spending money on keeping people.
Is a low score on one area worse than a low total?
Usually, yes, and that is why the panel leads with the weakest area rather than the total out of 120. A team scoring evenly in the middle is a general condition and it tends to improve when the business does. One collapsed area propping up three sound ones is a specific piece of work with a specific fix, and it is normally the thing making your week harder. The count of statements below your own average tells you which of the two you are looking at, so read that before you read the score.
Is anything I type here stored?
No. The whole calculation runs in your browser and there is no database behind this page, so no score, salary or resignation is recorded anywhere. If you want the longer write-up by email you can ask for it below, and only at that point is an address captured — the answer on this page is never held back for it. You can also print the result or save it as a PDF from the button on the result panel, which drops the site navigation and keeps the scores, the working and the starting point.