A position is what somebody says they want. An interest is why they want it, and the two come apart far more often than they hold together. A functional manager who says an engineer is unavailable until April has stated a position. The interest underneath might be a commitment of their own in March, a worry about losing the person permanently, or the plain fact that nobody asked early enough for them to plan around it. Only one of those is really about April.
Positions are what negotiations get stuck on, because there is nothing to do with two incompatible positions except split the difference or win. Interests can be traded, since two parties rarely want the same thing for the same reason, and a good agreement is usually one that costs each side something the other values more.
Four positions and what sat underneath them
| What they said | The interest underneath | What that opened up |
|---|---|---|
| The date is the thirtieth of June and it is fixed | A board meeting where a working demonstration has to exist, which only part of the scope touches | A June release of the part the board will see, with the remainder following in August |
| We need the full export feature in this release | One customer who threatened to leave, and one report their analyst cannot produce by hand | A single report shipped in a fortnight, which holds the customer while the feature is built properly |
| This defect has to block the release | Being named in the review of the last incident that shipped knowingly | A recorded decision carrying the sponsor's name, and a fix committed to the next release |
| The team can take no more scope this quarter | An unfunded support rota already eating a third of every week | Funding the support work separately, which returns the capacity the new scope needs |
The right hand column exists because interests can be traded and positions can only be split or won. Finding the interest takes asking why the thing is wanted, twice, in a tone that makes the answer safe to give. Where somebody declines to say, the fallback is to guess out loud and let them correct you, since a person who resists explaining themselves will usually correct a wrong reading of their motives.
Preparing, which is most of the work
Two things are worth knowing before the conversation starts.
The first is what you will do if there is no agreement. Your alternative sets the point below which walking away beats accepting, and it is a fact about your situation rather than a bargaining tactic. A project manager who has not worked it out negotiates as though every outcome except agreement is unthinkable, which is visible from the other side of the table and expensive. Improving that alternative before you negotiate, by lining up a contractor or a different sequence, does more for your position than any amount of firmness in the room.
The second is why the other party wants what they have asked for. That takes asking rather than guessing, and the question that opens it is some version of what would have to be true for this to work for you. People answer it, mostly, and the answer is frequently something you can supply cheaply.
What actually gets negotiated
Scope is negotiated whenever a stakeholder wants an addition and the honest answer is that it can happen if something else moves. Resources are negotiated with functional managers who own the people and carry commitments of their own. Schedule is negotiated with sponsors whose dates usually came from somewhere other than an estimate, often a market event or a board meeting. Priority is negotiated constantly, because two urgent things submitted by two directors is a question about the organisation's preferences that somebody has to put back to the organisation.
The last of those is the one project managers most often try to solve by working harder, which quietly converts a negotiation somebody should have had into overtime.
Negotiating inside an organisation
Negotiating internally differs from negotiating across a contract in one respect that changes everything. You will meet this person again, repeatedly, for years, and no signed document at the end makes the relationship irrelevant. A win extracted from a functional manager in March gets paid for in September, when you need a favour and the memory is entirely intact.
That makes the sustainable approach separating the people from the problem, being hard on the issue and easy on the person, and being visibly willing to lose small points. It also makes candour cheaper than it looks. Explaining the constraint you are actually under invites the other side to solve it, and a colleague who understands your problem will often find an option you had not.
The negotiations nobody calls negotiations
Almost none of this reaches a diary as a negotiation. Agreeing an acceptance criterion, asking for a week of a specialist's time, settling whether a defect ships or blocks the release, persuading a sponsor that a date is unavailable. Each has two parties, incompatible opening positions, interests underneath, and an alternative if no agreement is reached.
Noticing that they are negotiations is most of the benefit, because it changes what you do first. An unrecognised negotiation gets conducted by asserting a position and waiting to see whose is louder, which is how a project manager ends up losing an argument they had prepared nothing for. A recognised one starts with preparation and a question.