An ENTP man can construct the strongest available version of a position he does not hold, which is a real accomplishment and is socially identical to believing it. A seat requires that a room be able to find you. Twenty years of being the best argument in the building is twenty years of nobody knowing where you are.
In 2009 a broker walked us through a licensing deal on a series of patented perennials and everyone in that conference room wanted it, and Mercer Nyholm took it apart in about six minutes without ever saying he was against it.
He asked what the royalty was per liner. Then he asked what it was per finished pot at the garden centre. Then he asked what our margin was on the comparable unpatented series.
Then he said, “So the whole thing hinges on whether the tag sells the pot or the plant sells the pot.”
Corwin Halbach said the tag sells the pot.
“Might be right.” Mercer had a pen he was not using. “But if the tag sells the pot then the licensor owns the customer, not us, and in five years they can do this with anybody with a greenhouse and we’ve spent five years training our own accounts to ask for their name.”
Somebody asked whether he was saying we shouldn’t do it.
“I’m saying somebody should be able to answer that,” he said, and then he spent the next ten minutes making the case for the deal better than the broker had made it, and by the end of it two people in that room had changed their minds twice.
We did the deal. It ran nine years and it was fine and the licensor now sells that series to three of our competitors, which is precisely what he described, and nobody here has ever put those two facts together out loud except me, privately, a number of times.
Mercer came to Halbach in 2001 in the production office. He was twenty-five, with a horticulture degree and two years at a garden centre. He is a tall, fast-talking, extremely funny man who is genuinely delighted by a good objection, including one aimed at him. He reads three things at once. He will start a sentence, stop, say “no, that’s wrong,” and start again, and the second version is better, and he does this in front of customers. Twenty-five years and he has never walked out of a meeting angry, and I have watched several people walk out of meetings angry at him.
Halbach Greenhouses is in Wexford: nine acres under glass and poly, plus twenty acres of outdoor container. We grow young plants — plugs and rooted liners — for other growers, and finished material for about a hundred and forty independent garden centres and two regional chains. Corwin Halbach is the third generation.
I came here in 1993 selling. I ran sales from 2004 and I have been president since 2024.
Here is the sentence about Mercer Nyholm that this company has used for twenty years. It is affectionate, it is accurate, and I built the practice that made it true.
Nobody is better in that room. I could not tell you what he thinks.
The crop plan for the season after next is due to the growers in three weeks. It is the first one I sign as president. Mercer’s proposed version is on my desk and I have read it four times.
Corwin has a way of closing an argument that everybody here can do.
“Fine. Where does he actually land?”
Square Feet
This business runs on a decision you make two years before you find out.
A perennial we sell in the spring of 2028 gets stuck as a cutting in the summer of 2026. Propagation space is booked before that. Which means the crop plan — the document that says how many square feet of bench go to what — is a set of bets placed twenty-six months out on what a garden centre buyer in another state will want to put on a shelf, in a spring whose weather has not happened, for a customer who may not exist.
And you cannot hedge, because space is the constraint. Every square foot you give to one thing is taken from something else. No version of the plan lets you keep your options open. A greenhouse will not let you hold two possibilities; it makes you pick one, early, and then it grows whatever you picked whether you were right or not.
That is the whole reason a crop plan is a strange document to write and a stranger one to argue about. Everything else in this company can be discussed. The plan has to be a number.
We grow a coleus that Corwin’s father put into the plan in 1978 and that has not been a serious item since about 2004. It occupies nine hundred square feet. Corwin will not drop it, and the reason he gives is that a grower two states east still orders it, and the reason everybody believes is that his father named it after his mother. Every year somebody proposes dropping it and every year the coleus is in the plan, and it will be in the plan I sign.
I take one week off a year, in the same rented cabin, and I have done that since 1997. My wife has stopped asking whether we might do something else.
The Other Side
What Mercer does is not contrarianism and it is not showing off, and calling it either is how this company avoided looking at it for two decades.
Whatever is on the table, the adjacent versions of it arrive to him unbidden — including the inverse. Somebody says the tag sells the pot, and instantly, without effort, he is also holding the plant sells the pot and both sell the pot to different buyers and neither sells the pot, the price sells the pot, all live, all at once.
And then the second thing happens, which is the part nobody credits. Every one of those gets tested. Does it hold together, does it explain the thing we already know, what would need to hold. Most of them die inside a second and you never hear about them. What comes out of his mouth is what survived, which is why the objection nobody had is almost always right.
The trouble is what it looks like from a chair on the other side of the table.
He can build the strongest available version of a position he does not hold. That is a real accomplishment, it is rare, and it is socially indistinguishable from believing it. So a room hears a man argue brilliantly for the deal and brilliantly against it inside the same quarter hour and comes away with the entirely reasonable conclusion that he is somewhere else, or nowhere, or wherever the argument is best that afternoon.
Somewhere adjacent to this there is a man who sorts by what is worth doing rather than by what stands up, and he leaves a meeting sounding like an advocate. This one leaves it sounding like a court.
Imelda Frayne ran our technical services from 2011 to 2016 and did the same thing in the same meetings. What got said about Mercer is nobody is better in that room. What got said about Imelda was that she was combative, and then that she was exhausting, and I heard both of those said and said nothing, and she is at a breeding company in the Netherlands now. What the identical performance costs a woman in that meeting, and how much earlier the room stops enjoying it, is set down on the facing page.
Twelve people at director level or above. Two are women.
Imelda came to see me on her last day. I asked what we could have done, which I thought at the time was a decent question to put to somebody.
“Put me in the room with him and let us go,” she said. “Twice a year. In front of Corwin.”
I said that sounded like a fight.
“That’s what you call it when it’s me,” she said.
The Test
Somewhere around 2007 I started taking things to Mercer before I took them anywhere else.
Not to get his opinion. To get them attacked. I would bring him a program, a price move, a new account structure, and I would say find the hole in this, and he would, in about four minutes, and then I would go and fix the hole and take the fixed version forward.
The most useful forty minutes I ever spent in this building was a container-programme proposal in 2011 that I was going to take to Corwin mid-week and took to Mercer instead.
He read it standing up.
“What’s the freight assumption?”
I said we’d move it on our own trucks.
“On what days?”
I said whatever days the routes ran.
“The routes run loaded in the spring.” He turned a page. “You’re proposing to add a hundred and forty stops in the eight weeks a year when there is no truck. So either you buy trucks, which isn’t in here, or you’re shipping common carrier, which is a dollar-ten a unit you haven’t costed, and at a dollar-ten this whole thing is a wash.”
I said I’d fix it.
“It’s fixable,” he said. “Do it on a summer programme and it’s the best idea anybody’s had here in two years.”
We ran it as a summer programme. It is still running.
It is the single most valuable management practice I have ever had and I would recommend it to anybody. It saved this company an amount of money I could not calculate. Corwin picked it up from me. Gunnar Boe picked it up from Corwin. By about 2012 it was simply how things worked here: you take it to Mercer first.
Which means that for something like eighteen years, in every consequential conversation at this company, Mercer Nyholm was, by appointment, the opposition.
Not because he opposed things. Because I put him there, deliberately, as a function, and then everybody in this building formed an impression of a man they had only ever encountered in that function.
I did not see the shape of that until this winter. What I had built was a machine that produced the sentence I could not tell you what he thinks, and then I went and used the sentence.
Gunnar put it to me at the time as a compliment to the arrangement.
“Best thing about Mercer is he’ll take any side.”
I said that was true.
“Makes him useless for the top job, mind.”
I said that was also true, and neither of us examined the join between those two sentences, and we were standing about ten feet from the cabinet.
Now Do the Other Side
In 2015 Mercer came to my office about the chain business.
We had two regional chains taking about a fifth of finished volume. The chains pay late, they return, they set the retail price, and they had started asking for exclusivity on two of our better series. Mercer’s view — and I want to be careful here, because for once he arrived with a view — was that we should give the chains notice, take the volume hit for two years, and rebuild the independent base with the series they could not get anywhere else.
He had numbers. He had a two-year cash bridge. He had a list of forty independents he thought would take more.
He was not testing. He came in and said, in the first fifteen seconds, “I think we should do this and I’ve thought about it for a year.”
And I said the thing I always said, because it was the practice, because it worked, because it was a compliment in the currency this company used.
“Great. Now do the other side.”
And he did it. Of course he did it. He did it for twenty minutes and it was the best case for staying with the chains that anybody has ever made in this building — the volume, the cash timing, the risk of a two-year hole, the fact that four of my salespeople would leave. It was so good that Corwin, who came in halfway through, assumed that was his position.
We stayed with the chains.
That was eleven years ago. We are still with them, they are now a quarter of volume, the exclusivity happened in 2019, and last year one of them took a series to a competitor in Michigan and we lost the account for two seasons and got it back on worse terms.
He never brought it again. He has raised the chain question in meetings since, from various sides, entertainingly. He has never again walked into my office and said I think we should do this.
I did not notice that for a decade. I noticed it this winter, and the first thing I noticed was that I could not remember the last time he had told me what he thought without my having to work out whether he meant it.
Where Does He Land
The general manager job came open in 2018 and it went to Talbot Ness.
Talbot is good. He is organised, he is reliable, and when Talbot tells a customer something in the spring it is still true in the autumn, which in this industry is worth a great deal and which I am not being sarcastic about.
The conversation about Mercer took, in total, perhaps two minutes.
Corwin said what he says. Where does he actually land. And I said — this is the sentence I have gone over — that Mercer was the best thinker in the building and that I did not know how you would put him in front of a hundred and forty garden centres, because he would tell a customer the truth about our own product in the morning and the truth about a competitor’s after lunch, and both would be accurate and the customer would come away unsure who to buy from.
That is a real problem and it is a fair thing to say and I have not stopped believing it.
It is also the whole and complete description of a man being disqualified for the thing he is best at, by the person who had spent a decade assigning it to him.
Talbot came to see me a few weeks after it was announced, which he did not have to do.
“You know he’d have been better at the product side of it than I’ll be.”
I said he was probably right and that the job was not the product side.
“No,” Talbot said. “But it’s most of it.”
Mercer congratulated Talbot and meant it. He has worked for Talbot since and there has never been a bad word. About six months after, in the shipping barn, he said one thing to me about it.
“I’d have been all right at it.”
I said I knew.
“That’s not what I’m asking,” he said, and picked up a cart, and I have thought about what he was asking approximately once a month since.
Sunniva Toth asked me in her second year what Mercer’s position on the chains was.
I said that was a good question.
“I’ve heard him say four different things about it.”
I said that was Mercer.
She said, “Right,” and wrote something down, and I never asked what, and I have wondered since whether it was the four things or the fact that I had not answered her.
Twenty Plans
The crop plan is submitted by product and signed by the president.
Which means that every year since 2006, Mercer Nyholm has put his name on a document that allocates a finite quantity of square feet to a finite list of items, with numbers, in a format that cannot hold two positions at once.
Twenty years of them are in a cabinet in the office because we are required to keep them for the seed and cutting contracts.
I pulled all of them in the winter. It took two evenings and a table, and the reason I went and got them was not this. I was reconstructing a contract history for the seed company and the plans were where the contract quantities live.
His plan and the plan we actually planted are different documents in every year but three. And the differences are not scattered. He shifted space out of the chain-facing series and into the independent-facing series starting in 2013 and he did it again every single year afterward, in increasing quantity, and every year the plan we planted moved it back.
He proposed dropping the licensed perennial series in 2016, 2017, and 2018, three years before it went to our competitors.
He put square footage against a native perennial program in 2011, which we did not do, and which is now the fastest-growing category in this business and which we entered in 2022, badly.
Seventeen substantive calls over twenty years. He was right on thirteen of them, wrong on two, and on the other two it is not knowable.
I checked the two he was wrong about, because I wanted them to be bigger than they are. One was a shrub programme that a virus decision killed for the whole industry. The other was a colour.
He has been in exactly one place since 2013. He has said so, in writing, with numbers, once a year, in the only document this company produces that does not permit an argument.
And I have signed off on the alternative every time, and I have spent those same years saying that nobody could tell where he landed.
I took the 2013 plan out and read the cover memo, which is four lines and which he writes every year and which nobody reads. It says what changed from last year and why. The 2013 one says the chain business is buying our shelf position back from us a season at a time.
That is not a man testing an idea. That is a sentence with a date on it.
The Report
After 2018 Corwin wanted to do something for him, which was genuine, and what a company does when it wants to do something for somebody and has no seat to give is hire a coach.
The engagement was six sessions with a woman named Sabra Kwan and it cost twelve thousand dollars. The intake included one of the sixteen-box questionnaires: four letters and a short profile.
The report went to Corwin and to me. Mercer got a verbal debrief.
I want that sentence to sit on its own, because I did not think about it once in 2019 and I have thought about it a good deal since. We paid for a coach for a man, and the written product of that engagement was delivered to his employer and not to him, and everybody involved including me thought that was normal.
His page says ENTP.
The profile is a paragraph about quick thinking and enjoying intellectual challenge, which reads like something out of a magazine and which is what allows a person to decide the whole thing is nonsense and skip the appendix. The appendix is where the material is. It says four letters are shorthand for rank: which habit is out in front, which one it hands to, and which two sit at the back where a man has to go and fetch them, which takes time he does not usually have. The ranking for this one is set out under the type’s own entry, from the sixteen the instrument was cut out of.
Out in front is the arrival of the alternative. Put anything on the table and its other versions are simply there, the inverse included, unsent-for.
The second is a private test of whether a thing holds together — not whether it is good, not whether it is worth doing, whether it stands. That is the filter, it runs fast, and it kills most of what the first one produces before anybody hears it. It is also what lets him assemble the strongest form of something he does not believe, because assembling the strongest form is a coherence problem and coherence is what that habit does.
The pair at the back are what a general manager is made of.
One is the live read of what a room needs from him — the sense that this particular room, at four o’clock, with a decision due, does not want an examination, and the further sense of what a virtuoso performance costs the standing of the person whose proposal is being examined. He does not have that. He has never had it. He is not unkind; he is genuinely surprised, afterward, that somebody was upset.
The other is the retrievable record of his own past positions. That is the faculty which would have allowed him to say, in that shipping barn, I have said the same thing about the chains every year since 2013 and it is in the cabinet. He could not say it because he does not carry it. He does not know he has been consistent. He has never gone and looked.
So the sentence I built is true in the only way anybody was checking. He cannot be located by conversation, because conversation is where he tests. He could be located in about two evenings by anybody willing to open a cabinet, and in twenty years nobody was, including him.
The plain part, since I received a document about a colleague that he has not read. If your employer pays for coaching, ask in writing before the first session who receives the notes and the final report and what becomes of them afterward, because development engagements routinely generate documents that go into decisions you never see, and the time to settle that is before rather than after. And being asked to argue against your own proposal is a facilitation technique, not an evaluation of your position, and if the two are being conflated in anybody’s account of you it is fair to say so plainly and in writing. None of it belongs in a conversation about somebody’s temperament. The people for it are your manager, then whoever the company keeps for complaints, then a lawyer.
Sabra’s report has three development recommendations. The second one is that Mercer should work on demonstrating greater consistency of message.
We paid twelve thousand dollars for a recommendation that he stop doing the thing we hired him to do. I read that in 2019 and considered it fair.
Sabra called me at the end of the engagement to close it out.
I asked how it had gone.
“He’s the easiest client I’ve had all year,” she said. “He’ll take any note you give him.”
I said that sounded promising.
“It isn’t,” she said. “Nothing sticks to a man who agrees with you.”
Three Weeks
His plan for the season after next takes eleven hundred square feet out of the chain-facing series and puts it into natives and into two things we do not currently grow.
Talbot’s version, which is the version that will go to the growers if I do nothing, moves it back.
The cabinet knocked two things loose.
The licensing meeting in 2009. That has been a story of mine for years — the day Mercer argued both sides of a deal so well that two people changed their minds twice — and it is a funny story and I have got a laugh with it at a trade show. What he actually did in that room was state, in one sentence, the precise mechanism by which that deal would end — the licensor owns the customer — and then, because nobody in the room could evaluate it, do the honest thing and build the opposing case as strongly as he could so that we would be choosing rather than agreeing. The sentence was the position. The twenty minutes afterward was service.
Then my office in 2015, and now do the other side. I have carried that as good management and I taught it to three people. What it was, that particular time, was a man arriving with the only unhedged advocacy of his career and being asked, by the person he had come to, to dismantle it — and being so good at it that the dismantling became the company’s position for a decade.
What the report left out. There is a dispute with his own brother over their mother’s house since 2020 and will not discuss it; he is genuinely bad at holidays and disappears into a phone; and there is a former employee of this place, let go in 2014, and he has never once said the name, and he is the only person here who has never told that story.
I have his plan and Talbot’s plan side by side and the difference between them is eleven hundred square feet, which is the entire subject of every argument this company has ever had about it, and which nobody has ever had to say out loud because it has never been on a piece of paper next to itself.
I asked Sunniva Toth, who has been in product with him for three years and takes notes on everything, what she thought the biggest disagreement in this company is.
She said she didn’t think there was one.
The people in these essays are composites drawn from long observation. The theory is not.