Interests, Not Positions

“We cannot go below 95 percent fill rate” is a position.

The interest behind it might be that the plant manager was blamed for a stockout two years ago and has no intention of being blamed again. It might be that a contract carries a service-level penalty nobody has read recently. It might be that 95 was in a board deck and walking it back requires a conversation they do not want to have.

Three interests, three different resolutions, and none of them addressed by demonstrating that 94 percent is optimal.

Positions defend the rule that protects their holders

To the person holding one, and frequently to everyone else in the room, a position is indistinguishable from a constraint. Enter “98 percent fill” and “inventory down 20 percent” into a model as constraints and the model reports that the problem is infeasible, or feasible only under assumptions nobody believes.

A position is usually the shortest available description of what its holder stands to lose if the rule changes. The planner’s insistence on fill defends the pad that keeps angry calls from reaching her manager. Dealer operations’ insistence that every order be honored defends the expedite path that lets a region escalate its way to parts. Finance’s insistence on a uniform cut defends a number already promised upward. Each position protects an arrangement, and frequently that arrangement is a piece of the incumbent rule.

That is also why positions harden as the challenger becomes more concrete. A position is loss aversion expressed as a number: its holder can see exactly what the challenger puts at risk and can only estimate what it would deliver. What moves a position is an alternative that protects what the position was defending.

When positions deadlock, the decision is still made, by default, in favor of the rule already running. The meeting argues about your model and the incumbent fires again on Monday.

Positions, interests, and alternatives are three different objects

Bargaining as though they were one is how rooms get stuck.

A position is a demand: 98 percent fill, 20 percent down, honor every order. A position is not an alternative. It is a constraint somebody is trying to put on the ranking.

An interest is what the position stands in for: stranded jobs, cash, regional peace. Interests belong in the objective, in units of consequence.

An alternative is a commitment the owner could actually make: keep the current rule, take the cut, run classes at two centers. Alternatives belong on the menu.

The failure is to negotiate positions. Finance says 20, operations says 98, and the room settles at 10 and 96: two demands stated slightly more quietly, and a rule slightly wrong for every class of item. The useful sequence is to hold the positions long enough to translate them, invent alternatives that serve the interests underneath, and then rank those, together with what each party will do if the room produces nothing.

You cannot rank a class-based rule against “98 percent”. You can rank it against a blanket cut, the status quo, and finance taking the cut anyway.

Ask what the number protects

A guess about someone’s interest that is never checked is speculation presented as analysis. Ask, and ask in a way that offers to serve what lies beneath the number rather than challenging it.

The useful form is hypothetical and specific. If I could get you fewer stranded jobs without the 98 percent line-fill number, would you take it? If we could meet the cash target without touching the parts that stop vehicles, is that the same result for you, or a worse one? Neither question asks anybody to abandon anything in public.

The answers fall into three groups. Sometimes the interest is stated plainly and the position dissolves into it. Sometimes the answer is a second position, which usually means the person is repeating a number they were given and do not own, and the owner is further upstream. And sometimes the answer is a firm refusal delivered with some heat, which almost always means a commitment behind the number that the person is not free to explain.

Two conditions make the question work. Ask in private, before the meeting; in front of colleagues it is an attack. And the other person has to feel understood first, which technical people skip: they hear 98 percent, recognize a position, and move straight to the better alternative while the other person is still thinking about the last dealer with a vehicle on the lift. Restate the concern in ordinary words and wait: you are not defending a percentage, you are defending the week a water pump is missing and a dealer is on the phone. A correction teaches you something. Agreement gives you permission to proceed.

A planner who pads a minimum is protecting herself from that call, not sabotaging your model. Rank the alternatives as they will be run from her chair, including the pad, or you are ranking a company that does not exist.

The constraint test

Sometimes the position is real, and an analyst who treats every constraint as negotiable is as unhelpful as one who treats every position as physics. The test is whether whatever stands behind the number can be shown.

Ask what happens the day after the number is missed, and who finds out. If the answer is a penalty, it is economics: a contractual service level with a penalty schedule is a term to price and trade against holding cost, not a wall. If the answer is a covenant breach, a regulator, or a public promise already made, treat it as a constraint this cycle and rank the alternatives that remain. If the answer is that somebody will be annoyed, it is a position, and the interest behind it belongs in the ranking instead.

When finance’s 20 percent turns out to be a real cash event, say that the choice has changed rather than continuing to argue for a class-based rule on its merits. The decision is now how to take a required cut with the least damage, and the class analysis becomes the evidence for which items should absorb it.

The map is part of the model

Who decides, who can veto, who executes, who bears the downside, who receives the upside, and what each will do if no agreement is reached: this map is constraint data, not office politics. A rule that finance will not fund is infeasible. A rule that dealers will bypass is infeasible in the only sense that matters.

At Meridian Motors the map is short. Priya Shah owns the choice. Finance can force a cash event, and information technology can delay the parameter change for a cycle. Planners run the rule and dealer operations can defeat it in the field. A class-based rule that protects water pumps shifts pain onto the cheap class and onto anyone who wanted a uniform cut, and the write-up should say so. A write-up addressed only to the owner models a company with one person in it.

When interests genuinely conflict, write down who bears the cost, how much, and for how long, and put the trade in front of the owner. That case, and the fallbacks that arrive with it, belong to the next entry.

Being right is not enough

This does not claim that merit is irrelevant. It claims that correctness is one term in a product, and a product with a zero in it is zero. Whenever you bring a recommendation into a room, people are judging four things at once. The first three accumulate credit and the fourth spends it.

Competence is whether you appear to know what you are talking about. It is almost always already sufficient by the time you are invited to the meeting. Nobody is declining your recommendation because they doubt you can compute a holding cost.

Reliability is whether the thing you said would arrive on Thursday arrived on Thursday, and whether you flagged the problem yourself before somebody else found it. It accumulates only by repetition, which is why people who think influence happens during presentations undervalue it.

Safety is whether people feel able to tell you something awkward. Everything else on this page depends on it, because the interest underneath a position is usually slightly embarrassing: a manager who does not want another angry call, a director blamed publicly the last time service slipped. If nobody has ever told you something in confidence, you are not being told the interests, and your model of the decision is the one a stranger would build.

Self-orientation works against you. It is how much of your attention is visibly on your model, your framework, and whether you are seen to be right. Whatever competence, reliability, and safety you have accumulated are discounted by it. The first three take years. This one you can change in the next meeting, by saying that the recommendation is close, that a colleague’s objection is the right one, or that the number being asked about is the one you are least sure of.

Contempt is the specific failure to guard against

Technical people fall into it by temperament. Telling a room that 98 percent is arbitrary wins the exchange and loses the meeting. Telling them that 98 stands in for stranded jobs, and that here is a rule aimed at the jobs instead, is still a disagreement, but about something both of you want. The wants already on the table are a freeze somebody can defend, a cash number that survives review, and a Saturday that does not ring. Your elegance is not on that list.

A five-minute test

Take the two or three numbers that are currently boxing in your analysis. For each, write the position, the interest you believe it stands for, and the answer to “what happens the day after it is missed, and who finds out?” Mark each as price it, constraint this cycle, or position. If you cannot fill the interest column without guessing, you have a conversation to hold before the meeting, not a model to extend.