Negotiation Language

Learning outcomes

A recruiter offers you a salary. You want more, you know you are worth more, and you have one sentence to respond. You can feel the temptation to either cave (“that sounds fine, thank you”) or to plant a flag and defend it (“I need ninety thousand, that is my number”). Both feel like the only two options, and both are weak: the first gives away money you could have had, the second turns a conversation into a standoff where someone has to lose. The same trap shows up when you push back on a deadline, a project scope, or a vendor contract. The skill that gets you out of it is not toughness and it is not charm. It is a way of talking that treats the deal as a problem you and the other side solve together, and this page teaches that language.

After studying this page, you can:

• Separate the person from the problem, so you can be hard on the issue and easy on the human being across the table.
• Find the interest hiding under a stated position by asking why, and use it to unlock a deal that pure position-trading would miss.
• Invent options that expand the deal before you split it, and anchor a fair split on an objective standard rather than on willpower.
• Name your best alternative to a negotiated agreement, the source of your real power and your walk-away point, and prepare it before you ever speak.
• Use the core moves of the table (a reasoned anchor, the conditional trade, the soft decline, the probing question, and silence) on a salary and scope negotiation.
• Spot the five mistakes that sink negotiators, from positional bargaining to making it personal, and replace each with the better move.

Before we dive in

Most people picture negotiation as a contest of wills. Two sides arrive with a number, they push against each other, and the stronger nerve, the louder voice, or the better bluff wins the bigger slice. Under that picture, language is a weapon: you state your demand, you refuse to budge, you wear the other side down. This is what the researchers Roger Fisher and William Ury, in their book on getting to a yes, call positional bargaining, and the whole point of their work is that it is a bad way to reach good deals. It produces agreements nobody is happy with, it costs you the relationship, and it often fails to reach any agreement at all.

The problem they set out to solve is the one in the salary story: the felt choice between giving in to keep the peace and digging in to protect your interests. Fisher and Ury argue that this is a false choice, because it treats the negotiation as a battle over a fixed pie when it is usually a shared search for an arrangement that works for both sides. Reframe it that way and a different kind of language becomes available, language that attacks the problem instead of the person, that digs for the real reason behind each demand, and that grounds the outcome in something fairer than who can hold out longer. They call this principled negotiation, and it rests on four ideas plus one piece of preparation, which the rest of this page works through one at a time.

A few terms will recur, so meet them once. A position is what someone says they want, the specific demand on the table (“I need ninety thousand”). An interest is the underlying need or concern that the position is meant to serve (you want to cover rent, feel valued, not be underpaid versus peers). An anchor is the first number or proposal put down, which shapes the range the talk happens in. Your BATNA is your best alternative to a negotiated agreement, what you will do if this deal falls through, and it is the true measure of your power. Everything below is built from these few parts.

Mental Model: a search for a deal, not a fight to win

Here is the reframe the whole page turns on. The wrong model, the one most people carry, is that a negotiation is a tug of war: there is a rope, you both pull, and every inch the other side gains is an inch you lose. Under that model the language is adversarial by design. You hide information so the other side cannot use it, you treat their gain as your loss, and you measure success by how much you took. The trouble is that this model is usually false, and acting on a false model leaves value on the table for both of you.

The better model is that a negotiation is a joint problem to be solved, where you and the other side sit on the same side of the table and the deal is the thing across from you both. Their needs are not your enemy; they are data you need to build an arrangement that holds. The classic illustration is two people fighting over a single orange. Position against position, the only fair split is to cut it in half, and each walks away with half of what they wanted. But ask why each one wants the orange, and it turns out one wants the peel to bake with and the other wants the juice to drink. The interests do not conflict at all. One takes all the peel, the other takes all the juice, and both get everything they wanted from a single orange. That gain was invisible at the level of positions and obvious at the level of interests, and finding it required talking to the other person as a partner in a puzzle, not an opponent in a fight.

Breaking it down

1. Separate the people from the problem

Start with the move that makes every other move possible. Negotiations involve two things at once: the substance you are trying to settle (the salary, the scope, the terms) and the relationship between the people settling it. The first mistake is to let the two tangle, so that an attack on the problem feels like an attack on the person, and the person defends themselves instead of helping you solve it. When your manager hears “this deadline is impossible,” they may hear “you set a stupid deadline,” and now they are protecting their judgment rather than fixing the schedule.

The fix in language is to be hard on the problem and soft on the person, and to put the two of you on the same side facing it. You attack the issue plainly while signaling respect for the human being. Compare “you are being unreasonable about the timeline” with “I want to hit this date too, and I am worried the current scope will not fit it, can we look at it together?” The second names the same problem just as directly, but it frames you and your manager as allies against a shared obstacle, not as opponents. The wording that does this work is mostly about pronouns and stance: “we” and “the problem” instead of “you” and “your demand,” and a sentence that states your concern about the issue rather than a verdict on the person.

This is also where you handle emotion. When the other side gets defensive or heated, the move is to acknowledge the feeling without conceding the substance: “I can see this is frustrating, and I want to find something that works for you, so help me understand what is driving the concern.” You separate the emotional charge from the problem so the problem stays solvable. The deeper machinery of staying composed and keeping a relationship intact when the topic is tense belongs to difficult conversations; here the point is narrower, that you cannot solve the problem until you have stopped fighting the person.

2. Focus on interests, not positions

This is the heart of the method, and the orange split is its emblem. A position is the specific thing someone demands; an interest is the underlying need that the demand is supposed to satisfy. People arrive at the table with positions because a position feels concrete and defensible, but positions are where negotiations get stuck, because two positions can flatly contradict each other while the interests beneath them sit comfortably side by side. The skill is to drop below the stated position and reach the interest, because that is the level where deals open up.

You reach it by asking why. Not as an interrogation, but as genuine curiosity about what the other side is actually trying to achieve. “Help me understand what is driving the timeline” turns a fixed date into a reason behind the date, and the reason is something you can often satisfy another way. Suppose a client insists on a feature shipping by the end of the month, and you insist it cannot be done well in that time. Position against position, you are stuck. Ask why the end of the month matters and you might learn it is tied to a board demo, in which case a polished prototype of just that feature, with the rest following later, serves the real interest perfectly while relieving the impossible deadline. The positions clashed; the interests did not.

Read the diagram top to bottom: the two positions meet in a deadlock at the top, but following each one down with the question why reaches interests that do not conflict, and those interests combine into a deal neither position could see. The lesson is that whenever a negotiation feels stuck, you are almost certainly stuck at the position level, and the way out is down, toward the interests, with the question why.

3. Invent options for mutual gain

Once you are working with interests instead of positions, a new move becomes possible: before you divide the deal, expand it. Negotiators who skip this step jump straight to splitting a fixed pie, haggling over a single number as if the only question were who gets which slice. The richer move is to invent several possible arrangements first, looking for ones that serve both sides’ interests at once, the way taking the peel and the juice served both orange-seekers fully. You are searching for value that neither of you would find while staring at a single number.

The language here is the language of joint brainstorming, and its key feature is that it floats ideas without committing to them. You explicitly separate inventing options from deciding among them, so that proposing something does not lock you in. “What if we tried…” and “one option might be…” and “just brainstorming here, suppose we…” all open an idea up for inspection without either side having to defend or accept it yet. In a scope negotiation, instead of fighting over whether a feature is in or out, you invent: ship a smaller version now and the rest next quarter; trade this feature for dropping a lower-priority one; bring in a contractor to make the original date possible. Each is an option that might serve both interests better than the original yes-or-no fight, and naming several before choosing keeps the pie growing before anyone starts cutting.

4. Use objective criteria

Inventing options expands the deal, but at some point you still have to settle on terms, and here a contest of wills creeps back in. If the only argument for your number is that you want it, and the only argument for theirs is that they want it, you are back to a tug of war decided by stubbornness. The fourth idea defuses this: agree the result on some objective standard, a fair principle that neither side controls, rather than on raw will. You shift the question from “what do you demand” to “what is fair, by what measure.”

The objective criterion can be market data, an independent standard, precedent, or a principle both sides accept as fair. In a salary talk, the criterion is the market rate for the role, the level, and the location: “for this role and level, market data puts the range between this figure and that one, and given my experience I am asking toward the top of it.” Now the number is not your willpower against the recruiter’s budget; it is anchored to a standard outside both of you, and to push back the recruiter has to argue with the market rather than with you. The same move grounds a scope or deadline talk in past project velocity, an estimate from a neutral standard, or what comparable contracts charge. The language is “let us base this on…” and “the standard here is…” and “what would be a fair way to decide this,” which reframes the disagreement as a shared search for the right answer rather than a clash of demands.

Read each row as one tool and the right column as the job it does. Notice that none of these phrases is aggressive, and none is a giveaway either; each does real work while keeping you and the other side searching together. The next sections take the four ideas you now have and add the one piece of preparation that gives them teeth.

5. Know your BATNA, your real source of power

You can run every move above flawlessly and still get a bad deal if you do not know one thing before you start: what you will do if this negotiation produces no agreement at all. Fisher and Ury call this your BATNA, your best alternative to a negotiated agreement, and they argue it is the true source of your power in any negotiation. Your power does not come from how badly you want the deal or how tough you sound; it comes from how good your alternative is. If you have a strong alternative, you can walk away, and the ability to walk away is what lets you hold firm on what matters. If you have no alternative, you are negotiating with a gun to your own head, and the other side can feel it.

This is why preparation, not performance, wins negotiations. Before a salary talk, your BATNA is your current job or another offer in hand, and knowing it tells you the exact point below which you should walk away rather than accept. Before a vendor negotiation, your BATNA is the next-best supplier and their price. The work is to make your alternative concrete and, where you can, to improve it before you ever sit down, because a better alternative is a stronger negotiating position with no change in tactics at all. A second job offer does not just give you a fallback; it changes how you can talk in the first negotiation, because now you can genuinely decline.

6. The language at the table

The four ideas and your BATNA are the strategy; this section is the tactics, the specific moves you make in the moment. The first is the anchor, the first number or proposal put on the table. Anchoring matters because the first number quietly shapes the whole range the talk happens in: people adjust from the anchor rather than ignoring it, so an opening figure pulls the final figure toward itself. This is why you usually want to put down a reasoned first number rather than wait, and why the anchor should be ambitious but justifiable. An anchor with no reason behind it is just a demand and invites a counter-demand; an anchor tied to a standard (“market data puts this at the top of the band, which is where I am”) sets a high but defensible starting point that the other side has to negotiate down from.

The second tactic is the conditional trade, and it is the single most important habit at the table: never give a concession away for free, always exchange it. The structure is “I can do X if you can do Y.” When you want to give something up, attach a condition that gets you something back: “I can pull the date in by a week if you can drop the reporting feature,” not “fine, I will pull the date in.” The free concession trains the other side to push for more and gets you nothing; the conditional trade gets you a return and signals that everything is part of a package. The third tactic is the soft decline, declining without killing the deal. A flat no ends the conversation; the move is to reject the specific offer while keeping the door open: “that does not quite work for me, but here is what could,” which says no to the terms and yes to continuing.

The fourth tactic is the probing question, which is how you reach interests and keep the other side talking: “help me understand,” “what would it take to make this work,” “what is driving that.” These pull out the information you need and, crucially, signal that you are trying to solve the problem together rather than just win. The fifth tactic is the most overlooked: silence. After you make an offer or ask a question, stop talking and let the silence sit. The instinct is to fill it, often by softening your position or talking yourself down before the other side has even responded. Resisting that, holding the silence, puts the next move on them and frequently draws out a concession or a real answer that your nervous chatter would have buried.

7. Salary and scope, end to end

Now watch all of it run on the two negotiations a software engineer faces most: salary and scope. Take salary first. Before anything else you do the preparation: you research the market rate for the role and level (your objective criterion), and you make your BATNA concrete (your current role or another offer). At the table, the first move is often to get the other side to anchor, by asking for the range: “what is the band for this role?” If you must anchor first, you anchor high and reasoned, tied to the market data. When the offer comes in low, you do not accept the first offer and you do not just demand more; you counter on the criterion: “based on the market rate for this level, and given my experience, I was expecting something closer to the top of the band.” Then you go quiet and let them respond.

The move that separates a strong salary negotiator from a weak one is negotiating the whole package, not a single number. Base salary is one variable among many: there is also the signing bonus, equity, the annual bonus target, vacation, a remote-work arrangement, a title, a review date. Fixating on base salary alone throws away the conditional trade across the whole package: if they cannot move on base, you trade (“I understand base is fixed at this level; could we close the gap with a signing bonus or an earlier review?”). You expand the deal before dividing it, exactly as inventing options prescribes, and you find value that the single-number fight would have missed. Scope works the same way. A request to add a feature is a negotiation: you find the interest behind it (why this, why now), you invent options (a smaller version, a later phase, a trade against a lower-priority item), you anchor the conversation on capacity and past velocity as objective criteria, and you trade conditionally (“I can take this on if we push the analytics work to next sprint”). In both cases the structure is identical, only the variables differ.

Calibrated directness and the repeated game

Two final adjustments separate competent negotiators from natural ones. The first is calibrated directness. How hard you push and how much you soften depends on the relationship and the culture, and getting it wrong cuts both ways: too blunt and you damage a relationship you need, too soft and your real position never lands and you get walked over. The fix is not a fixed level of politeness but a setting you adjust to the person and the stakes, firm on the substance, warm on the relationship. The full machinery of dialing your force up and down, of stating a firm position without sounding aggressive and of softening without sounding weak, is the subject of hedging and stance, and it is the exact skill that lets you be hard on the problem and easy on the person at the same time.

The second adjustment is the repeated game. Most negotiations at work are not one-off encounters with a stranger you will never see again; they are with a manager, a colleague, a client, or a vendor you will deal with again next quarter. That changes the optimal strategy completely. In a one-shot deal you might be tempted to extract every last concession, but in a repeated relationship the deal you strike today sets the tone for every deal after it, and a counterpart who feels squeezed will remember. So you negotiate for an outcome both sides can live with and walk away from feeling respected, because preserving the relationship is itself a long-term win. This does not mean going soft; it means you make your case on the merits, trade fairly, and let the other side keep their dignity, so that next time they come to the table trusting you. The connection between persuading on the substance and keeping people on your side over the long run is developed further in persuasive writing; in negotiation the same truth holds, that the agreement worth having is one the other side would still say yes to if they saw everything you knew.

Common mistakes

The first mistake is positional bargaining: arriving with a fixed demand and pushing on it, position against position, instead of digging for interests. It locks the talk into a tug of war over a single number, misses the deals that only appear at the interest level (the orange split), and tends to damage the relationship even when it works. The fix is to drop below positions to interests by asking why, then invent options that serve both sides before you split anything.

The second mistake is having no BATNA, walking in without knowing what you will do if the deal falls through. With no alternative you cannot credibly hold any line, because both sides know you have to say yes, and the other side negotiates accordingly. The fix is to work out and, where you can, improve your best alternative before you ever speak, so you know your walk-away point and can push from real strength.

The third mistake is giving concessions for free, conceding a point because the other side pushed, without getting anything back. Each free concession weakens your position and trains the other side to push harder. The fix is the conditional trade: every concession is exchanged, “I can do X if you can do Y,” so each move you give gets a move in return.

The fourth mistake is accepting the first offer. A first offer is an opening position, usually set with room to move, and accepting it both leaves value on the table and signals you would have gone further. The fix is to counter, anchored on an objective criterion, and to treat the first number as the start of the conversation rather than the end of it.

The fifth mistake is making it personal, letting an attack on the problem become an attack on the person, or treating the other side as an opponent to beat rather than a partner in a deal. It puts the other person on the defensive, so they protect themselves instead of solving the problem, and it spends a relationship you will likely need again. The fix is to separate the people from the problem: hard on the issue, soft on the human, the two of you on the same side of the table facing the deal together.

Mastery Questions