Negotiating hard and keeping the relationship

The idea

You can be hard on the number and easy on the person, but only if you sequence it. State the constraint before the demand. Give the reason. Give notice in proportion to the size of the change. Never let a cut arrive first as an email.

The ask itself is not what damages a supplier relationship. A 40% volume cut is a 40% volume cut whether you open with it or close with it, and the number you reach is roughly the same either way. What differs is what the ask cost you: whether they could still plan around it, whether they know it was your budget rather than their quality, and whether they will tell you the truth next time you need something from them.

That last part is the whole reason to care. In a relationship you will renegotiate every year, goodwill is not sentiment. It is the thing that converts into information, into capacity when you are short, and into a better opening position in round four.

How it works

The sequence below is the whole method. None of the steps make the ask smaller. They change what the ask costs.

1. Decide whether this is a repeated game. One transaction and never again, and you can play the number. A supplier you will renegotiate every year for a decade is the other game, and in that game the process is the product.
2. Give notice in proportion to the change. A price review needs a week. Half their volume needs a quarter, because a quarter is roughly what it takes them to go and find replacement work. Notice is not a courtesy. It is the difference between a cut they can absorb and a cut that lands on their payroll.
3. State the constraint before the demand. "Our program budget was cut 40% in June" is checkable, and it arrives before the number rather than after it. A reason offered after the ask is heard as justification, which is worth much less.
4. Say the ask is not a verdict on them. Out loud, in one sentence: this is our budget, not your quality. Suppliers assume every cut is a scorecard until you tell them otherwise, and a supplier who thinks they have been graded starts protecting themselves instead of solving with you.
5. Make the ask plainly, once, with the number in it. Vagueness is not kindness. "We need to look at volumes" makes them guess low and hear it twice. Say 40%, say when, and stop talking.
6. Say what remains and what would bring it back. A cut with no floor sounds like the beginning of an exit. "You keep the 60%, it is committed for 24 months, and the line comes back if the program is refunded" is a different conversation from the same 40%.
7. Ask what it does to them, and mean it. This is not softness. It is the cheapest information you will get all year: which of their costs are fixed, what their real capacity floor is, whether they can re-time it. You only get the honest version if the first five steps went well.
8. Trade something cheap for you and valuable for them. Term, volume certainty, a faster payment cycle, a forecast they can actually plan against, fewer emergency changes. These often cost you very little and are worth real money to a supplier with a working capital problem.
9. Never let the cut arrive first as an email. Size the channel to the size of the change. Half the volume is said in person, by the person who owns the decision, before any system sends a purchase order that says it for you.

The arithmetic behind the demo

Every number in the interactive comes from this table, so you can audit it. The ask moves the commercial meter by the same 26 points every time. Only its relationship cost changes.

opening position                number 40        relationship 78

run one, delivered cold
  make the ask             +26  ->  66      -22  ->  56
  name the alternative     +16  ->  82      -10  ->  46
  ask about their side     +11  ->  93       +9  ->  55
                                =======           =======
                                     93                55

run two, in order
  state the constraint      +5  ->  45       +6  ->  84
  make the ask             +26  ->  71      -13  ->  71
  offer something back     +14  ->  85      +15  ->  82  (held at the ceiling)
                                =======           =======
                                     85                82

the ask, three ways
  cold, no warning and no reason        +26        -22
  with one of the two                   +26        -13
  after notice AND a reason             +26         -6

Two of the moves are not fixed at all. What a supplier tells you, and what they will do with a trade, both scale with the meter, which is the mechanism the whole page is about:

ask about their side  ->  2 + round(20 x relationship / 100)
offer something back  ->  4 + round(14 x relationship / 100)

at relationship 46:  side = 2 + 9  = 11
at relationship 71:  back = 4 + 10 = 14

And a cold ask sets a ceiling. Once you have surprised someone, nothing you say in the same meeting takes the relationship back above 65. With one of notice or a reason, the ceiling is 82. With both, there is no ceiling, because there was no surprise to repair.

The repair, which is the part people skip

A hard round is not over when the meeting ends. Three things actually rebuild the meter, and none of them happen in the room.

• The debrief you offer. Come back in a week, on your own initiative, and walk them through how the decision was made and what would change it. Offering it is most of the value; whether they take it matters less.
• The first small commitment you keep afterwards. The 30 day payment terms you promised, the forecast you said you would send, the engineering answer you owed them. Keep the small one fast. It is the only evidence that the hard round was about the number and not about them.
• The thing you concede next time. Relationships are scored across rounds, not inside one. If you took everything this year, give something visible next year, and say why. A supplier who can predict that pattern will bring you problems early instead of hiding them.

The internal half

Half of these negotiations are lost inside your own building, to a stakeholder who wants a harder line than the relationship can carry. Arguing tone with them does not work, because tone is not their currency. Price it in theirs.

• Supply risk, in their units. "If they cut a shift, our lead time goes from 6 weeks to 11, and the launch date moves." That is a schedule conversation, which they will engage with.
• Switching cost, fully loaded. Requalification, tooling transfer, first article inspection, the engineering hours, the scrap during ramp, and the months. Put a number and a date on it. Most aggressive positions soften the moment the alternative has a real price.
• What we spend the goodwill on instead. Goodwill is a budget, not a mood. If we spend it on 3% this quarter, we do not have it for the expedite we will need in the fourth.

When to use it

Six situations these questions actually contain, and the move each one wants.

• A volume cut of half in person, with the floor and the way back Deliver it face to face, name what remains, and name what would bring the volume back. The trade-off is that it costs you a trip and an uncomfortable hour, and there is no version of this that is comfortable. An email saves the hour and costs the decade.
• A supplier heading into financial trouble restructure terms, not price Squeezing a supplier who is already thin is a supply risk you are choosing to own, and you own it silently until the day they miss. Faster payment, longer commitment, smaller order quantities and a firmer forecast can be worth more to them than the discount you were going to ask for, and cost you less. The limit: if they are genuinely failing, the honest move is a second source with notice, not a rescue you cannot fund.
• You are most of their revenue phase it, and say the phasing out loud That much leverage converts into dependence, and using all of it once is how you inherit their capacity problem. Phase the change, tell them the phasing up front so they can replace the volume, and watch their concentration as carefully as they should be watching yours. The trade-off is real: you leave money on the table this year to keep a supplier solvent enough to serve you next year.
• A competitive bid you are required to run publish the policy, the timing and the criteria Trust here comes from the process, not the outcome. Tell the incumbent the policy exists, when the bid opens, and exactly what it will be scored on, before it starts. An incumbent who loses a process they understood usually stays reachable. One who finds out from a rejection letter does not. You cannot promise them the result, and you should not try.
• A friend on the other side of the table disclose, then run a documented process The appearance is the exposure, whatever the decision turns out to be. Declare the relationship in writing to your manager and to compliance, put the scoring criteria on paper before you see the bids, and where the spend is material, let someone else hold the pen. The limitation to accept: doing this properly may cost you the friendship, and doing it improperly costs you the job.
• Repeated quality escapes, and requalifying is expensive pressure on the system, with a date Put the weight on containment, root cause, corrective action, a measurement you both agree on, and a date, rather than on the person who happened to take your call. Escalate by level, not by volume: your director to their director is pressure, shouting is only noise. The trade-off is patience, and the backstop is that the date is real and there is a stated consequence when it passes.

Watch out for

• Notice delivered after the decision. Telling a supplier on Friday about a cut that was signed on Tuesday is not notice, and they can usually tell. It reads as managing them rather than respecting them, and it is worse than saying nothing, because it converts a commercial decision into a question about your honesty. If you genuinely could not tell them earlier, say that, and say why.
• Leading with the alternative. Opening with "we have another source at 11% under you" before you have said what the problem is turns a fact into a threat, and threats get discounted. The same sentence, placed after the constraint and the ask, is information they can act on. In the demo it is the difference between paying 20 points of relationship for 8 points of number and paying 10 for 16.
• Confusing hard on the number with hard on the person. Refusing to name a figure, letting silence do the work, implying their quality is the reason when it is not, going around them to their boss on the first pass. None of these make the number better. They make the ask feel like a verdict, and a supplier defending their reputation stops solving your problem and starts building a file.
• Spending all your leverage the first time you have it. When you are 60% of a supplier's revenue you can extract almost anything once. Then their margin goes, then their best people go, then their capacity goes, and the next expedite you need is not there. Leverage that big is a position to manage, not a button to press.
• Treating repair as something that happens in the room. A warm close after a cold ask does not undo it, which is why the demo caps how far the meter climbs back inside the same meeting. What actually repairs it is the debrief you offer a week later, the first small promise you keep, and the visible concession you make next round. If you are not going to do those, do not perform the warmth.

Worked example

The question is usually some version of this: "Tell me about a time you had to take a lot of cost or volume out of a strategic supplier. How did you handle it?" A weak answer describes the tactics used to win the number. A strong one describes the order of the conversation, and then prices what it cost.

Here is the shape. Nine year supplier, roughly 15% of your category spend, and you are about 55% of the revenue on the line that makes your part. Your program budget is cut 40% in June, effective in the third quarter. You call their commercial lead the same week the decision is made, before anything is in a system, and say the volumes are going to move, the number is not final, and you will have it in ten days. That call moves no money at all. What it does is let them keep a shift busy on other work rather than discover the gap when a purchase order changes.

Ten days later you go and see them. You state the constraint first, with the actual board decision and the date, then say plainly that this is your budget and not their quality, because they will assume the opposite. Then the number: 40% off the line from the third quarter. Then the floor, immediately: the remaining 60% is committed for 24 months, and the volume returns if the program is refunded. Then you stop and ask what a 40% cut does to them, and you listen to the answer, which is where you learn that a third quarter start costs them a shift but a fourth quarter start does not, because another program ramps in October.

So you phase it: 20% in the third quarter, the rest in the fourth. That costs your budget one quarter of savings and costs them no layoffs. In exchange you ask for the thing that is cheap for them and valuable for you, a firmer price hold across the 24 months, and you give them the thing that is cheap for you and valuable for them, 30 day payment terms instead of 60. You did not get the full 40% in the third quarter. You got the cut, the price hold, a supplier who now tells you about their capacity problems early, and an opening position next year that is not a repair job.

Then the part that makes the answer sound senior rather than tidy: name what it cost and what you would watch. You gave up one quarter of savings, which you had to defend internally against a stakeholder who wanted the whole cut immediately, and you defended it in supply risk and switching cost rather than in tone. You also concentrated more spend on a supplier where you are already 55% of a line, so the standing action is to qualify a second source deliberately and openly over the next year, with them knowing, rather than as a lever produced in the next negotiation.

Check yourself