Who Must Be Persuaded First

In any organization a small number of people determine whether a change is taken seriously, and their agreement is worth more than a majority.

They are not necessarily senior. They are the people whose behaviour others read as evidence, and identifying them accurately is most of the work.

Three categories

The credible sceptic. Somebody respected, not reflexively negative, known to say what they think. If they engage with a change, everybody concludes it survived scrutiny. If they visibly do not, everybody concludes there is a reason.

The operational hinge. Somebody whose daily work the change actually runs through. Their cooperation is not symbolic; without it the thing does not physically happen.

The informal authority. The person others consult before forming a view. Frequently long-tenured, frequently not senior, and their position is invisible on any chart. In a small organization there are usually two, and everybody except the leadership can name them.

The useful question for identifying the third is not who is influential. It is who people ask. Put that question to somebody a level or two down and the answer arrives immediately.

Go to the sceptic first

The instinct is to build momentum among the willing and approach the difficult person once the thing has support. This is backwards for three reasons.

Their objection is most useful while the design can still change. Raised after the proposal is fixed, a valid objection becomes an obstacle rather than an improvement.

Approaching them late is visible and reads as vote-counting. A person consulted after the outcome is settled knows exactly what they are being asked for.

And a proposal that has survived the strongest objection is a different proposal. It arrives at everybody else already tested, and can be presented as such.

What you are asking for is not agreement

Requiring people to agree sets a standard many cannot honestly meet and produces performed enthusiasm from those who try.

The workable ask is narrower: implement it properly, do not undermine it, and tell me if it is not working. That can be met by somebody who thinks the change is wrong, and it is all the change actually requires.

Stating it that way also removes the incentive to pretend, which preserves the information you will need at month four. A person who was permitted to remain sceptical will tell you when it is failing. One who had to profess belief will not.

The person who cannot be persuaded

Occasionally somebody important will not move, and the programme cannot wait indefinitely.

Establish first which situation you are in. Somebody with an unanswered objection is not refusing; they are waiting for an answer. Somebody who has lost something real is not obstructing; they are absorbing a cost that has not been acknowledged. Both are addressable and both are commonly misread as refusal.

Where the objection has been answered, the cost named, and the position holds, the conversation is direct and private: the decision stands, I need you to implement it, and if you cannot, tell me. A capable person occasionally says they cannot, which is honest and leads somewhere.

What must not happen is proceeding around them while pretending otherwise. Everybody can see it, and it teaches that quiet obstruction is survivable.

The hinge cannot be persuaded by argument

A distinction worth holding: the sceptic is persuaded with reasoning, and the operational hinge frequently is not.

Their question is not whether the change is a good idea. It is whether it works on Tuesday with the actual caseload, and no argument answers that. Only a demonstration does.

Which is why the pilot matters more than the presentation for this group, and why running it in a genuinely representative setting matters. A hinge who has watched it work in circumstances resembling their own has been convinced. One who has been shown a favourable pilot from a soft environment has been shown nothing.

In a small organization

Below about forty people the three categories collapse into two or three named individuals, and everybody knows who they are.

This makes the work simpler and the stakes higher. There is no diffusion: if the long-serving supervisor is visibly unconvinced, the change does not happen, and no amount of organizational communication compensates.

It also means the conversations are relational rather than procedural. In an organization where the same twelve people will still be working together in five years, how somebody is brought into a change matters as much as whether they were, and a person managed rather than consulted will remember it long after the change itself is routine.