Negotiation position generation

The job is a one-page internal brief the negotiator can still explain at the table: BATNA, a target range from a named should-cost, a walk-away, an opening ask inside that band, and the concessions you will actually trade. If you cannot point to evidence for a line, it does not belong on the page.

It is not a script or a savings claim. It is not a mandate until the category owner signs the walk-away. It is never something you email to the supplier.

SAP Ariba, Coupa, Keelvar, and Jaggaer are the class of sourcing suites where the event, bid files, and award recommendation already live. Pull quotes, ranks, and history from that file. Those platforms are not ranked here. None of them writes your BATNA or your reservation price. The position sits on top of the event, in a place suppliers cannot see.

A model that invents the supplier's cost structure is not preparing you. Public filings can tell you whether a plant is closing, whether one customer dominates their book, or whether they flagged going-concern risk. They do not tell you this SKU's margin. Do not print a "room to move" percentage.

BATNA is what you do if this conversation fails. Name it in operational terms: which bidder takes how much volume, after what qualification, on what lead time, at what landed cost including split freight and dual tooling. "We can always go to market" is not a BATNA.

Build it from competitive pressure analysis: the second-place bid that can actually run the spec, the line-level gaps you will press, and the volume the runner-up can absorb this quarter. If the second plant is not qualified, the BATNA is delay plus a trial, not an instant switch.

Single-source risk flagging tells you when the BATNA is a fiction. One qualified plant, one region, one tool: your walk-away is delay, a spec change, or a dual-source program you have not started. Write that. Do not invent a phantom second source so the target looks brave.

Switching cost belongs on the page: qualification weeks, safety stock, tooling, customer notification if a branded pack changes. A cheaper second bid that cannot ship for ninety days is not cheaper on day one.

The target range is the low-to-high delivered should-cost for this spec, this Incoterm, this ship-to, from market price benchmarking. Name the series, the last-PO cohort rules, and freight. If you cannot open those sources, you do not have a target. You have a wish.

The opening ask sits inside that band, typically at or near the low end you can still defend with the stack. It is not a number below the low end "to leave room." A target below should-cost with no BATNA is a request that the supplier subsidize you. They will refuse, or they will win and miss on quality, or they will reopen the price in the next cycle.

The walk-away is the worst deal you will sign: usually the BATNA's total landed cost, or the high end of should-cost plus switching, whichever you would actually choose. Write it as a price and as a structure (split award, delay, re-run). A walk-away with no alternative behind it is a bluff. Bluffs leak.

Do not take last year's PO as the target because the incumbent already lives there. Last year may have been high. Do not take the incumbent's first quote as the ceiling. Do not let the model estimate conversion cost and SG&A from a 10-K and then subtract a made-up margin. That is inventing the supplier's P&L.

Multi-criteria decision matrix already ranked this event on the published weights. Do not reopen quality, ESG, or risk as a free price chip unless the category owner is changing the award. If B won on total score, close B or close a split. Do not talk A down while pretending the ranking did not happen.

A concession is a change you will put in the contract, not a discount you invent because the clock is running.

List the trades in advance, each with what you give and what you require:

• Term length for a price at the mid of should-cost, not for goodwill.
• Volume commitment or SKU rationalization, such as dropping a specialty width, in exchange for holding the opening on the core SKU.
• Payment terms or Incoterm that finance has already approved.
• Forecast visibility or a firm call-off window, if operations will honor it.

If operations will not honor a volume commit, do not offer it. If finance will not move off current terms, do not dangle faster payment. Contract term pre-population is where those trades should already exist as clause options. The position page points at the clauses you will open. It does not invent legal language in the meeting.

Price last, after the structure is clear. An unearned discount ladder with no BATNA and no should-cost is not a plan. It is a countdown to last year's number.

This walkthrough is illustrative, not a measured result. It claims no savings rate.

A packaging buyer is awarding machine-grade stretch film, 20 micron, 500 mm, DAP to one Midwest DC, after a three-bidder RFQ. The incumbent has the lane. Bidder 2 ran a plant trial last quarter and can take a stated share of volume on the existing wrapper setup. Bidder 3 quoted a sharp unit price from a regional line that has not been trialed on this film and this machine.

The should-cost file names a film-grade LLDPE series, twelve months of DAP POs for this spec with expedites stripped, and contracted truckload to the DC. The file shows a delivered band: low, mid, high. The opening on the page is the low end of that band. The target is the mid. The walk-away is Bidder 2's landed cost for a split, including extra SKU handling and dual-lane freight, not Bidder 3's untried unit price.

BATNA: award a split to Bidder 2 at the trialed share, keep the rest with the incumbent at walk-away, or delay for a Bidder 3 trial if operations will hold safety stock. Two plants are already qualified, so sole-source language does not apply.

Concessions the buyer will actually trade: a 24-month term if the incumbent meets the mid; drop a specialty 750 mm SKU if they hold the opening on 500 mm; payment-term changes only if treasury already signed them. They will not offer a volume commit operations cannot forecast.

What they almost did: the draft inferred a gross margin from the incumbent's last 10-K and set a target below the should-cost low, with no BATNA because Bidder 2's volume cap was ignored. A one-pager went to the incumbent so they could "see the math." The incumbent treated the opening as the ceiling and the invented margin as an insult. The useful page stayed internal, named Bidder 2 as the alternative, and kept every dollar inside the should-cost band.

Treat the generated brief as a starting argument. The category owner signs the walk-away. The negotiator can move inside the signed band. They cannot move the band because a model sounded confident.

• A target below should-cost with no BATNA asks for a subsidy. Write the BATNA or raise the target.
• Inventing the supplier's cost structure is not analysis. Strike any line that asserts conversion cost, plant utilization, or SKU margin unless you have a teardown or your own should-cost stack. Filings are context, not a P&L for this SKU.
• Treating the draft as the mandate leaves you with a memo. If the owner has not signed the walk-away, you do not have a position.

Keep the page in the event file, supplier-blind. Do not paste it into a sourcing-suite message, a transparency email, or the RFQ addendum. The numbers stay inside.

The job is finished when a negotiator can open one page, name the alternative, point at the should-cost sources, state the opening and the walk-away, list two or three trades they will sign, and walk into the room without a second deck.