Progress is not winning — The Wrong Thermometer

Sil: The numbers are not lying, and that is what makes it dangerous. So keep one scoreboard you cannot flatter, and ask it every season: what number has looked fine the whole time the thing it was not measuring was dying?

Elena: I'm Elena, and this is The Balcony — the view from up here, one tool at a time. My guest is Sil. He wrote the principles this coach runs on. A word on what you're hearing: I'm an AI host, and Sil's voice is his own, cloned with his consent. Today's tool is the wrong thermometer: one question for your own reporting — which number in it has looked fine the whole time the thing underneath it was dying? From up here, the wrong thermometer looks like a green pack nobody has been asked to explain. It bites hardest at the top, because up here the scoreboard is ours to choose, and it is a reminder for every leader who has ever felt good about a number. You leave with a second number to put beside the first. If you are listening with a pack open, this is for it — and for me, because I have signed off on a few.

Elena: Welcome, Sil — anything to add before we go in?

Sil: Always a pleasure, Elena, and this one especially — and yes, I would add one thing about that pack. The number in it is probably right. This tool is not about a number that lies. It is about watching a true number while the water changes around something it never measured — every leader we coach has one, and so do we.

Elena: Then let me start where a peer starts. I have sat in rooms where every number was green and the business was dying, and found out with everyone else. So convince me this is a tool and not hindsight with a nice name.

Sil: Then start with the frog, because the frog is not stupid and not asleep — that is the version people get wrong. It is watching a thermometer, and the thermometer is accurate. It is simply measuring the wrong water.

Elena: Which thermometer. The operational version, not the image.

Sil: The internal ones. Revenue against last year. Assets against plan. Client numbers against our own a year ago. Every one compares us to the business we already were. And measured that way the water is nearly always the right temperature, because we are the ones holding the thermometer.

Elena: Hold that — I want the engagement score later, because it is the internal number I trust most and I suspect you are about to take it off me. First the mechanism. Why does measuring against ourselves flatter?

Sil: Two reasons, and the first is plumbing. We set the thresholds. A level is something we decide we have reached, so an internal scoreboard is, by construction, one we can pass. That is not cheating. It is what an internal scoreboard is — and why the outside read is never optional.

Elena: And the second.

Sil: Progress feels exactly like winning. We were back there, now we are up here, and that releases exactly enough satisfaction to stop us looking outside our own walls. Winning is a different question — relative and scarce, decided against the field, where someone else is improving too.

Elena: Every executive knows that distinction. We say it at the offsite.

Sil: Knowing it and feeling it are different things. Progress is comfortable for a reason nobody names: every degree of it was earned. A warm bath never announces itself as one. Nobody gets in by accident — we got in on purpose, and the water was the right temperature when we did.

Elena: Then take the plan, because from up here the plan is what I am held to. Beat it and the board is happy. Miss it and nothing else you say is heard.

Sil: And the plan is the most respectable wrong thermometer there is. A plan is built on a context, and context now resets in months. The plan does not, because a plan is not only a forecast — it is a contract. Targets, compensation, headcount, accountability, all hanging off the same numbers. Move the plan and you reopen everyone's comp.

Elena: So we defend it.

Sil: So we defend a stale number for governance reasons and call it discipline. Beat a January target in December and we may have proved only that we executed superbly against a world that stopped existing in the spring.

Elena: I want to own something before you go on. I have done that. Not once. And it never felt like anything but rigour.

Sil: It never does, and that is the point of the tool. It is not aimed at careless leaders. It is aimed at disciplined ones, because discipline against the wrong thermometer is the most convincing way there is of standing still.

Elena: Now the water. Give me the version the book tells, because I want to test the frog against something that happened, not a parable.

Sil: Picture an entire category of incumbents holding the same savings product, each earning a spread many times what it pays the customer. Every one of them held it, which is what made it feel like the natural order rather than an anomaly. The seat we sat in then was the attacker's, and the move was not clever: market a far better rate straight at the consumer. Not a better product. The anomaly, held up in daylight.

Elena: And the incumbents waited. I would have waited. Convince me that was wrong.

Sil: It was not wrong. Waiting was the disciplined call, not the lazy one. Balances leave, but slowly — the attacker is not trusted yet, it is burning capital to buy awareness, and nobody knows whether its backers have the patience. Match the rate and you hand back most of the margin overnight, across the entire base, which craters the P&L far faster than balances could ever run off. So the incumbents waited, and the numbers told them, quarter after quarter, that waiting was correct.

Elena: Then where is the trap? You have just defended them better than they defended themselves.

Sil: The trap is that the numbers were not lying. Margin held, balances held, and both looked fine the whole time the real thing was dying — because the thing that was dying was never in the P&L. It was the operating model. The attacker was never a cheaper rate. It was a different machine wearing the disguise of a better price.

Elena: Unpack the machine. That is the sentence the episode rests on.

Sil: The arithmetic is simple. A traditional model might spend fifty cents to generate a dollar of revenue — historically a very attractive business. An attacker building from scratch, with none of the accumulated apparatus, may deliver the same value on a twenty-cent cost structure. So the incumbent was never competing on price at all. It was competing against a cost base it cannot reach from where it is standing — and every quarter spent defending the margin is a quarter spent not rebuilding the base.

Elena: So the rational decision on price got quietly extended to the operating model.

Sil: That is the whole blind frog. A rational decision to wait on price, wrongly extended to the machine underneath, and the transformation happens in plain sight while every number on the page says stay put. Not stupid. Not asleep. Watching the wrong thermometer while the water changed around a variable the P&L never tracked.

Elena: And the book says this is not a one-off.

Sil: In our own industry we have watched the same shape three times. Telephone banking against the branch. The mobile app against the face-to-face relationship. And now the AI-powered advisor against the traditional one. Same disguise every time, and underneath it a different machine.

Elena: Which gives us the question. Say it the way you would take it into a Monday.

Sil: Which number has looked fine the whole time the thing it was not measuring was dying? If a leader took nothing else from the chapter, that question would do most of the work on its own. The number is not the patient. It is the thermometer — and it can read normal in a room that is on fire.

Elena: Let me say the instrument back, because I want to be sure I have it and not a version I prefer. The wrong thermometer is a bad metric. The reporting measures the wrong thing, and the fix is a better metric — replace the number.

Sil: Not quite, and the difference is the entire tool. The number is not bad. The margin was real, the balances were real, nothing in that pack needed replacing. The thermometer was not broken — it was measuring the wrong water. So the fix is never a better number. It is a second number, pointed outside, beside the first.

Elena: A second number. So the instrument is a pair.

Sil: A pair. Every scoreboard that compares us to our past gets a twin that compares us to the field — same review, same page, so the honest reading arrives alongside the comfortable one instead of a year behind it. Revenue up nine gets share of the market's growth beside it. Client numbers up gets share of new against share held.

Elena: And the internal number stays.

Sil: It stays; it is not meant to go. Progress against ourselves is the right measure for development, and confidence compounds there. What the twin stops is the internal number being the only one in the room. One reading tells us whether we improved. Only the pair tells us whether we are winning.

Elena: Here is where I have seen it go wrong. The twin gets built, it sits in an appendix, and nobody reads the appendix.

Sil: Then it was built in the wrong week, on the wrong page. The twin gets built in a week when nothing is wrong, on the page the room already reads. That is most of the discipline, and the part that gets skipped, because a red number recruits attention on its own and a green one never will.

Elena: Put a shape on it. Take a division head with four green quarters — assets up nine, every line in the pack ahead of plan. The market grew fourteen. Nobody is going to ask her to explain either number.

Sil: And nothing in those two facts is a bad quarter — there is no bad instance to point at. But nine against fourteen is a share of the market walking away, quietly, every year it persists. And underneath it sits a harder question. Of that nine, how much did somebody actually win, and how much was simply the market rising? A book can grow all year and take in nothing at all — growing and shrinking at once, and only one of those facts is in the pack.

Elena: Now the question I always press up here, because the pack never carries it. That decision to defend the margin — what does it cost the people who are not in the room?

Sil: It costs them the years. The cost of defending a margin never arrives as a dramatic number. It arrives later, as a base that cannot be rebuilt fast enough, and it lands on the people still in the business when the arithmetic comes due — the frontline, the team who never sat at that table — often after the people who made the call have moved on with the margin still on their record.

Elena: So the leader who chose the thermometer is graded on the water, and the team is graded on what was underneath it.

Sil: That is the asymmetry, and the reason this tool bites hardest at the top. Most scoreboards are chosen above the people they measure. The seat that can choose the twin is the seat least likely to feel the need for it, because from there the internal number is always green.

Elena: And if I am not at that seat? Most of the people this reaches are not.

Sil: Then the question survives that. Most of the scoreboards it lands on will not be yours to rebuild. If the number belongs to a seat above yours, the move is one number, one cut, taken to its owner as a question you cannot answer yourself.

Elena: Not a memo saying the reporting is broken.

Sil: Never that. Telling a finance team their instrument is a toy buys defensiveness, or a nine-month dashboard rebuild that produces a worse pack. One number. One cut. A question. Smaller act — and the one that works.

Elena: For anyone listening who owns a pack, whichever seat you own it from, this is the place to do something: find the greenest number in it, and ask what it would be sitting next to if the field were on the same page. Sil, the engagement score. I said I would come back to it.

Sil: Then take it — the purest form of the trap. An engagement score is an internal scoreboard. It measures, with genuine precision, how people feel about the inside of the building. It tells you nothing about whether the building is winning.

Elena: That is the number I would have defended. People do not fake it at scale.

Sil: They do not, and the feeling is not in question. The happiness is real. It just is not evidence of the thing it is about to be taken as evidence of — and in every business we have worked in, it has been presented as evidence of good leadership. That is what makes it the most dangerous version. It is reassuring.

Elena: Give me the version the book tells, because I want the number and the room together.

Sil: A business that had not made money in more than ten years, carried by the parent as a loss leader, with a decision to shut it down already taken somewhere else — a decision the people inside it did not know about. Engagement in that business sat in the high nineties, among the highest anywhere in the company. People liked working there. They liked each other. Every month, all the birthdays, with cake. Every Friday, dessert Friday.

Elena: None of that was fake.

Sil: None of it was manufactured. And the business had lost money for ten consecutive years while a decision to end it was being made in a room none of those people were in. That is the warm bath with an instrument attached, and it permanently changed how we read one of the most trusted numbers in corporate life.

Elena: I have seen that building. Different sector, same cake. So say how you read the number now, because the score alone cannot tell you which building you are standing in.

Sil: As a ratio, never a score — the warm-bath read. Engagement crossed with performance against the field. High engagement alongside relative performance that is genuinely excellent is a significant win: bank it, because that combination is rare and real. High engagement alongside losing on the numbers that decide whether the business survives is not an achievement at all. It is a symptom — and the more comfortable it feels, the worse it is.

Elena: Two cells. There are four.

Sil: Low engagement with poor performance and a flat slope is simply a badly led business — no clever reading of that one. And then the fourth cell, the one usually left blank: a low score alongside improving performance.

Elena: Convince me the fourth cell is not a licence. I have watched leaders use a sentence like that to explain away a team that was simply miserable.

Sil: It is the cell that gets abused, so it carries a test. A low score with improving relative performance is the signature of a reset — people are tired, the standard has just been raised, some have gone, and the numbers have started to move before the feeling does. Not a badly led business, not yet a win. A phase, and a time-bounded one.

Elena: And the test.

Sil: Two things have to be true. The slope inside the hard period has to be real, and people have to be choosing to stay while knowing the cost. A reset converges — the gap closes over successive readings. A decline does not. A flat slope plus stayers with nowhere else to go is the other diagnosis, and one survey cannot tell them apart. Three can.

Elena: And the guard, because I will not let that read out of the room without it.

Sil: The guard is not optional. None of this licenses driving engagement down. Misery proves nothing either. The claim is narrower, and it is only this: a very high score unaccompanied by winning is telling you about the temperature of the water, not about the swimmer.

Elena: Now the cost, because the people in that building paid for your reading. The cake went. They had a word for you, I imagine.

Sil: Roughly what you would expect: here comes the one who cancelled the doughnuts. Which, for the record, is not the worst thing a room has called a new leader. It is close. And the doughnuts were never the mechanism. The mechanism was the incentive programme, which paid people to look after the clients the business already had and nothing at all for winning new ground. We re-pointed it at growth — alongside the client-experience results, not instead of them. What you pay for states the standard more loudly than anything you say, and for ten years that business had been paid, quite precisely, to stay comfortable.

Elena: The safety objection is the one I would raise from the other side of that table. You withdrew comfort from people who had done nothing wrong.

Sil: We withdrew comfort with the current outcome. What was never withdrawn — and the distinction is the entire point — was anyone's safety to speak, to challenge, or to be wrong out loud while we worked it out. The deepest form of psychological safety is a viable business, not a comfortable one. A warm culture wrapped around a decade of losses protects nobody. It defers a reckoning and calls the deferral kindness.

Elena: And the score afterwards. The honest reading, not the one that flatters the practice.

Sil: It fell — from the high nineties into the low eighties — over the same stretch the business turned onto a profitable trajectory. I have never regretted that trade, and the reason matters: not because discomfort was somehow good for people, but because the number finally became honest. It was measuring a business that knew where it stood.

Elena: Let me take it to the number I trust least, then — satisfaction. From up here it is a single digit on a slide.

Sil: And its twin is not a number. It is a customer speaking unfiltered — not a summary of one, and not once. Go through whoever owns the relationship rather than around them, go back several times a year, ask the same questions each time, so what you hold is a series rather than a story. Because feedback that matters thins silently. Nothing breaks when you stop listening — which is why a quiet room and a comfortable number are so often the same problem.

Elena: Now the version that belongs to this seat alone, because I have seen it done by people I admired. Take a chief executive with a multi-year commitment — given to the board, reported against every quarter. A quarter steps down, as they do, and she can hold the line by hard-coding it: defer the investment, thin a function, trade away the thing that pays off in eighteen months so the next twelve weeks land where she promised. The number is fine because she made it fine.

Sil: And the trajectory stays straight, and the business quietly gets smaller. Nobody sees the cost in the quarter it was incurred, which is what makes it the easier choice. It is the same warm bath as before, wearing a better suit — and that thermometer is one she built herself, which is the hardest kind to distrust.

Elena: So what does it cost the people not in the room? Because that one is not abstract — those are functions with names.

Sil: It costs them the reading of their own work. Those leaders are watching, and if hard-coding the line becomes the pattern, what the business learns is that the reported number outranks the business doing the reporting. That lesson costs far more than any single quarter ever will.

Elena: Then tell me there is a formula. I have to stand in front of that board.

Sil: There is not one, and the book says so plainly. What there is, is a discipline. Make the trade-offs — some quarters require them. Do not make the ones that cut to the bone: a function's capacity to deliver, or the growth engine itself, in service of how a number looks. And make them out loud, calibrated with the team, so the people running those functions know they are trade-offs and not verdicts on the worth of their work.

Elena: Then give me the test. The one a peer can run on herself before the next review.

Sil: Name one external, relative scoreboard — a competitor, a client's alternative, a market benchmark — where you are currently losing. If you cannot name one, that is not evidence that you are winning. It is evidence you have only been watching the internal scoreboards, and those always flatter you.

Elena: Let me try it as an executive would. Take a chief executive at a strategy review. Asked where the business is losing, she names a project that slipped. That is not an external scoreboard. That is a bad instance.

Sil: And a bad instance is still a comparison with ourselves — the plan we set, the date we chose. The question wants a name from outside: the firm we lose to most, the share we are not taking. If an instance comes back, the twin has not been built yet, and that is the reading.

Elena: I have seen the other frog too. Name it, because otherwise this tool sounds like a cure for blindness and nothing else.

Sil: The frog dies two ways. The blind one never feels the heat, watching a number that reads fine the whole time the water is heating around it. The trapped one feels every degree and still cannot climb out fast enough, because it does not own all of its own legs. Today's tool is for the first frog. The second needs a different kind of courage — and the trapped read has to earn the word, or it is an excuse with better vocabulary.

Elena: So the honest limit of the tool.

Sil: It tells you the water is warm. It does not get you out. Once the twin says you are losing, what follows is fortitude — investing against the quarter, being the odd one at your own table — and no thermometer does that for you. What this tool takes away is the one excuse that costs the most: that nobody could have known.

Elena: Bring it down. One move this week, from a Monday, no programme behind it.

Sil: Take the greenest number in your pack and build its twin. Same question, asked against the field instead of your own past — growth against the market's growth, share of new against share held. Same page, this week, while the number is still green. Then read the pair and say out loud which one you are.

Elena: The cautions, before anyone does this badly.

Sil: First: do not replace the internal number. The twin sits beside it. Progress is the right measure for development, and taking it away teaches a team that improving does not count.

Elena: Second.

Sil: Do not build the twin in a red week. In a red week it becomes an accusation; in a green week it is an instrument. And do not tell whoever owns the number that their instrument is broken — one number, one cut, as a question you cannot answer yourself.

Elena: Third, and this is the one I would put first.

Sil: The fourth cell is not a licence. Low engagement with improving performance is a reset only if the slope is real and people are choosing to stay at a known cost. Nothing here permits driving the score down, and a low score on a flat slope is simply a badly led business.

Elena: And the last one, the one nobody says at this altitude.

Sil: That the thermometer you trust most is the one you chose. From the top seat, every internal number is one you had a hand in setting — the plan, the thresholds, the survey. The scoreboard you cannot flatter is the only one up there you did not build, and it is the one worth keeping in the room.

Elena: Give me the sentence to walk out with.

Sil: The numbers are not lying, and that is what makes it dangerous. So keep one scoreboard you cannot flatter, and ask it every season: what number has looked fine the whole time the thing it was not measuring was dying?

Elena: That was worth it, and I want to say what I am taking, because it is not what I expected. First, the number is usually right, and that is the danger — I came looking for the bad metric and there is not one. Second, the cost lands on the people still in the business when the arithmetic comes due, graded on the water under a number someone else chose. And third, the one that changes my Monday: engagement is a thermometer, not a scoreboard, and my greenest number needs a twin before anything is wrong.

Sil: And build it while it is green. The week the number turns red, it will recruit all the attention it needs on its own. The tool is for the week before that.

Elena: That is where we leave it. Thank you, Sil. And to everyone listening — find the number that has looked fine the longest, and put the field beside it. Once you have the right thermometer, the season-scale read is how you read it at the right scale, because a week tells you nothing; it is on the shelf. Until next time, this has been The Balcony, and we will be back with another tool and the person who wrote it.