Duncan Reith

Honestly? Whether the chair likes them. I know that's not the answer anyone wants, and it's not the answer I'd have given at forty. But I've watched capable people fail because the relationship never took, and mediocre people survive three bad years because it did.

It's why I pay more attention to the coffee before the process than the process. Two people either find a way of talking to each other or they don't, and no scorecard has ever told me which it'll be.

The bit the machines do was never the hard bit in my world. Half my candidates aren't performing on the internet; they're running factories. The hard bit is knowing which plant director actually did the turnaround and which one stood near it. That isn't written down anywhere a model can read it.

So the research got cheaper and the judgement got dearer. Suits me fine.

Charisma. It gets people hired and it gets businesses into trouble, usually in that order. A factory doesn't care how good you are in a room.

Give me somebody the supervisors trust over somebody the analysts enjoy, every time.

Both, and the trick is knowing which one you're doing. In industrial there's a real version: if you've never run anything unionised, a heavily unionised site will eat you, and saying so isn't prejudice, it's arithmetic.

But when someone says "fit" and can't finish the sentence, that's not assessment. If you can't write the mismatch down in one line, there isn't one.

Somebody who has run something through a bad patch. Five years ago every brief wanted a growth story. Now boards ask me for people who have closed a site without losing the trust of the three they kept open.

Energy costs and reshoring did that. The heroic phase is over; the competent phase is here, and I prefer it.

They confuse the deputy with the successor. The deputy is the person who kept the machine running while the boss was away. The successor is the person who could change the machine. Sometimes it's the same person. Usually it isn't, and nobody's told the deputy.

No. An operations director in Rotherham who has run three sites for twenty years costs less than a strategy hire who has never stood on a shop floor. Explain that to me.

The market pays for polish and calls it potential. The gap is where my business lives, so I shouldn't complain.

When the answer's inside the building. If your best supervisor can name your next plant manager, promote them and save my fee. Some of the most useful conversations I have are the ones where I talk myself out of a mandate.

Sounds like bad business. It's the opposite: those companies come back for the searches that are real.

In my corner, less than anywhere, and the honest reason is upstream. My shortlists today are the engineering intakes of twenty-five years ago. You cannot fix in the boardroom what stopped happening in classrooms.

What we can do is stop making it worse: we bench every brief against the whole market, not the usual names, and we tell clients when their spec quietly excludes everyone who didn't have one particular career.

Seventy-thirty the company, and I'll die on that hill. I've watched the same managing director be brilliant under one owner and sunk by the next. Same man, same factory, different shareholder.

A plant manager with no degree, for a group operations role. The board nearly walked over it. He'd taken three failing sites and turned every one, and the only thing missing was a certificate from thirty years ago.

He's the group managing director now. The certificate still hasn't come up.

"Why are you still doing this?"

Because in 2011 I placed a man into a business everyone said was finished, and he rings me every Christmas Eve to tell me how the year went. Four hundred people still work there.

You don't get that feeling from many jobs. That's the answer, and nobody's ever asked me for it.