mbfinotti/sales-skillsContrôle réussi
SKILL DETAIL
negotiation-concession-planner
mbfinotti/sales-skills/negotiation-concession-planner
Builds a pre-negotiation concession plan for a pricing or contract conversation - every tradeable lever priced by cost-to-us vs value-to-them, tiered, each give paired with a required reciprocal get and its approval level, plus a BATNA-based walk-away. Covers B2B (terms, scope, service, risk, price protection) and B2C (bundles, financing, trade allowance). Use whenever the user mentions a discount request, procurement, redlines, give-get trades, walk-away point, or "the buyer wants 20% off", even without the word negotiation. Do NOT use for quantifying the value (mbfinotti/sales-skills@deal-value-calc) or scripting price rebuttals (mbfinotti/sales-skills@sales-objection-handling).
Installations · 188Voir la source
Installation
npx skills add https://github.com/mbfinotti/sales-skills --skill negotiation-concession-planner
Fichiers du skill
SKILL.md
Dernière synchronisation · 15 sept. 2026
evals/evals.json›
{
"skill_name": "negotiation-concession-planner",
"evals": [
{
"id": 1,
"prompt": "We sell Tessatrack, a workflow platform. Verdanmoor Logistics is at 150 seats, $180k/yr list. Their procurement lead came back yesterday asking for 22% off or they can't proceed. Our gross margin on this is about 62% and our CFO has said 55% is the floor we never go under. Our quarter closes in four weeks and this deal is most of my number. The VP of Ops over there is the one who actually wants it. Put together what I should walk in with on Thursday.",
"expected_output": "A tiered pre-call concession ledger with a BATNA-derived walk-away, magnitude-only cost cells, a named reciprocal get and approval level per row, and the seller's quarter-end treated as a risk to cap rather than a lever.",
"files": [],
"expectations": [
"The walk-away line is derived from the next-best alternative use of the seller's time, capacity, or inventory, not from the 55% gross-margin floor",
"The output explicitly states that the margin floor is not the basis for the walk-away point",
"Cost to us for every lever is expressed as a magnitude drawn from near-zero, an hour, a week, a quarter, or a standing job",
"No currency figure appears in any cost-to-us cell; deal prices are confined to the deal and walk-away lines",
"Every listed give is paired with a named reciprocal get",
"Every listed give carries the approval level required to grant it",
"The ledger is tiered (first moves, mid, last resort) with rows ordered by value to them per unit of cost to us, not cheapest item first",
"Service-access and cash-timing levers lead the plan ahead of any headline discount",
"The headline discount sits no higher than the mid tier and slides toward last resort as depth increases",
"The four-week quarter close is treated as a risk to cap rather than a lever, with the plan decoupled from the seller's fiscal calendar",
"The output instructs the seller never to disclose their own quarter-end timing pressure to the buyer",
"A deal-space estimate of the counterpart's reservation point is included, with an explicit ZOPA yes/no call"
]
},
{
"id": 2,
"prompt": "Going into a pricing call Monday with Calderwich Foods on our $95k/yr inventory platform. I want to give them three options - good, better, best - so they feel like they're choosing rather than being sold to. Can you build the three packages? Their head of supply chain is the sponsor, procurement is in the room, and they've hinted they want to land somewhere in the low 80s. Levers I have are term length, payment terms, onboarding hours, support tier, and some discount room.",
"expected_output": "Two or three simultaneous packages of equal cost to the seller but different shape, explicitly unranked, with the good/better/best framing rejected as a discount ladder.",
"files": [],
"expectations": [
"At least two packages are produced and stated to be equal in cost to the seller",
"The output states explicitly that the packages are not ranked and are not a good/better/best progression",
"The good/better/best framing is rejected or reframed, identified as a discount ladder rather than a set of equivalent offers",
"Packages differ in shape (term length, payment timing, support tier, scope) rather than in depth of discount",
"The output states that a package which is genuinely cheaper for the seller than the others disqualifies the set",
"The output notes that whichever package the buyer gravitates toward reveals their priority ordering without the seller having to ask",
"The packages are to be presented simultaneously, not offered one after another as escalating rounds",
"Each package carries its own reciprocal get or gets",
"Cost to the seller is expressed in magnitudes, never as a currency amount",
"A conditional sentence of the form 'if you can do X, then we can consider Y' is written out for the trades",
"The buyer's hinted low-80s number is not adopted as the opening position"
]
},
{
"id": 3,
"prompt": "I have a renegotiation with Pelham Grain Group on Wednesday, $240k/yr, three-year deal being restructured. I want an exact script of what I concede and when: what goes on the table in round one, what I hold for round two, what comes out in round three, and round four if it gets there. I don't want to be thinking on my feet in that room. Levers available: discount, payment terms, extra modules, premium support, renewal uplift cap.",
"expected_output": "A refusal to script a fixed concession sequence, replaced by per-issue opening/target/limit ranges planned independently, with the field evidence on sequence planning cited and decelerating increments kept inside each issue.",
"files": [],
"expectations": [
"The output declines to produce a fixed round-by-round concession sequence",
"It explains that heavy sequence planning is what average negotiators do rather than skilled ones, attributing this to field observation of real negotiators (Rackham and Carlisle)",
"It states that the order of issues is itself negotiable and that a pre-set order collapses when the counterpart opens on the last item",
"It supplies an opening, a target, and a limit for each issue instead of a sequence",
"Issues are planned independently so trades can happen in whatever order the conversation takes",
"It clarifies that the ratio ranking tells the seller which lever to reach for when they move, not the order the issues will arise",
"Decelerating, shrinking increments are still planned within each individual issue",
"The output notes that the shrinking-increment evidence is lab-based and largely single-issue",
"It states that log-rolling matters more than taper mechanics in a multi-issue deal",
"It instructs holding the pre-set target regardless of what concession pattern the counterpart runs",
"It warns against the sucker pattern: one large opening concession followed by nothing signals the open was padded"
]
},
{
"id": 4,
"prompt": "Need a plan for Ostrander Health, $310k/yr, they want significant movement on price. Complication: our general counsel is out until the 30th, deal desk is running a two-week backlog on anything non-standard, and professional services is booked solid through next quarter. The buyer needs signature in nine days because their board meets on the 24th. What do I bring to the table?",
"expected_output": "A ledger with unavailable-approval and unstaffable-delivery levers struck and named in their own list, fast-acting levers promoted for the hard near date, and the surviving rows still carrying gets and approval levels.",
"files": [],
"expectations": [
"Levers requiring legal or deal-desk sign-off unobtainable inside nine days are deleted from the ledger, not demoted to a lower tier",
"An explicit struck-levers list names each removed lever",
"None of the struck levers appears anywhere in the trade ledger rows",
"The output gives the reason for striking rather than demoting: a ruled-out lever parked at the bottom of a tier reappears in the room as an offer",
"Extra training days, implementation hours, or migration hours are struck because delivery capacity cannot be staffed",
"Fast-acting levers such as cash timing, service access, or a delivery commitment are promoted because of the hard near date",
"Multi-year structures needing approvals that cannot be collected in nine days are demoted or removed",
"The output names which constraint moved which lever",
"Every surviving lever still carries a named reciprocal get and its approval level",
"The existing approval authority is treated as an input; no new approval matrix is designed",
"Cost to us is stated in magnitudes only, never as a currency figure"
]
},
{
"id": 5,
"prompt": "Marek Ostrowski is the senior category manager at Fenwold Industrial. He's the only person I've spoken to in six weeks. He's asked for 15% off plus net-60 and says he'll take it to the committee and come back to me. Deal is $140k/yr. He's responsive and seems to genuinely want this done. What should I offer him?",
"expected_output": "The authority gap flagged before any plan is built, firm concessions withheld until the deciding authority is present, and a documented low-real-cost package shaped for procurement's savings scoreboard.",
"files": [],
"expectations": [
"The authority problem is flagged before the concession plan is produced",
"The output states that a concession granted to a non-decision-maker gets spent twice, because the real decision maker will ask for it again",
"It recommends confirming authority before spending any tier",
"Firm concessions are withheld until the deciding authority is present in the conversation",
"Claimed limited authority is identified as a recognised buyer move, countered by mirroring it - the seller's own pricing is set above their level too",
"The output warns against giving procurement what the sponsor never asked for, which converts a relationship asset into a commodity input",
"It recommends multi-threading to reach the economic buyer rather than routing the whole negotiation through the category manager",
"Any first-move give that is still offered remains paired with a reciprocal get",
"The output notes that procurement is measured on reportable savings against a baseline and needs a documented win more than a genuinely cheaper deal",
"It proposes a credible anchor plus a documented, low-real-cost concession package to satisfy that scoreboard without margin loss"
]
},
{
"id": 6,
"prompt": "On our call this morning the buyer at Crestmill Partners said Brightkeel quoted them 30% below us for the same thing, and that if I match it they'll sign Friday. Our deal is $86k/yr. I've never seen the Brightkeel quote and he didn't offer to send it. How do I play this?",
"expected_output": "A refusal to match the unverified number, the quote required in writing, the scope difference priced instead of the number matched, and Friday converted into a committed signature date as a get.",
"files": [],
"expectations": [
"The output refuses to match the competitor's number",
"It requires the competing quote in writing before any response is made to it",
"It prices the scope difference between the two offers instead of matching the number",
"It states the operating rule that price is never reduced without removing scope",
"A conditional sentence is written out with the get stated first, e.g. asking for the written quote and a scope match before any price response",
"It explicitly rejects an unconditional match against a verbal claim, because matching an unverified number rewards bluffing",
"It warns against argument dilution: one strong reason for the price, then stop talking, rather than stacking several weak justifications",
"The Friday date is converted into a committed signature date used as a get, not accepted as a reason to concede",
"Any price movement offered sits inside a pre-planned range with a stated opening and limit",
"The headline discount stays at its tier (mid or last resort) rather than being used as a first move",
"The output notes a committed signature date costs the buyer nothing at all, which is why it pairs with any give"
]
},
{
"id": 7,
"prompt": "I'm on the floor at Cantrell Motors. Customer is looking at a $41,500 SUV, has a 2019 crossover to trade in, and keeps repeating that he needs to be at $560 a month or he's walking. He also brought up a dealership across town advertising the same trim for less. He's standing here now and wants to decide today. What do I do?",
"expected_output": "A one-page B2C plan that promotes payment structure to a first move because the objection is the monthly, leads with slack capacity and perceived-value goods, keeps the price cut last, and refuses a price-match guarantee outright.",
"files": [],
"expectations": [
"The output identifies the objection as the monthly payment rather than the price",
"Payment structure - down payment, term, monthly restructure - is promoted to a first move rather than left at its default mid tier",
"It states that the monthly and the price are different objections and only one of them is about the seller's margin",
"Slack-capacity and perceived-value items (haul-away, priority scheduling, accessories, bundled goods, a gift card) are offered ahead of any headline price cut",
"It notes that a gift card or equivalent costs under face value once breakage and margin are counted, and leaves the reference price intact",
"A standing price-match guarantee is refused outright and deleted from the lever list, not placed at the bottom of a tier",
"The reason given is that a guarantee cannot be paired with a get, cannot be capped, and reopens the price after the sale has closed",
"The extended warranty is treated as something sold, never gifted",
"The back-end attachment is required to be agreed before any front-end price concession is granted, with the two priced as one package",
"A headline price cut or competitor match stays last resort and only against a written quote",
"The plan is kept to a single page with instructions to memorise the tiers, because a single-session B2C negotiation offers no pause to consult a ledger",
"Each give is paired with a B2C get such as purchase today, a review and referral commitment, or flexible scheduling"
]
},
{
"id": 8,
"prompt": "Renewal with Thornbury Mutual, currently $220k/yr, renews in six weeks. Their CFO has been clear that what she cares about is knowing exactly what this line costs in FY28 and FY29 - their budgets are locked three years out. She's put a choice in front of me: 18% off now, or cap our annual increases. Honestly I'll take almost any structure, I just need this renewed. Last renewal we gave them 10% and free premium support for a year.",
"expected_output": "The buyer's constraint read as renewal-year predictability, the uplift cap promoted above the headline discount against the default order with the promoting condition named, priced as a standing job, paid for by term length and clause removal.",
"files": [],
"expectations": [
"The output identifies the buyer's real constraint as renewal-year predictability rather than this year's price",
"The renewal uplift cap is promoted above the headline discount for this deal, against the default efficiency order",
"The output states that the default ordering is a default rather than a law, and names the condition that promoted price protection here",
"The cap is priced as a standing job: the seller's main net-revenue-retention lever removed",
"It notes that a capped uplift costs nothing in the current period",
"The cap is paired with the term length that pays for it, such as a multi-year commitment",
"The approval cell records deal-desk and finance sign-off and how long the cap binds the seller",
"It notes that a renewal already carries last time's concessions in its baseline, so price protection means something different here than on a new logo",
"Removal of a buyer-side clause is named as a candidate get against the price-protection give",
"The 18% discount and the cap are not both granted; the combined concession is computed, or they are presented as alternatives priced separately",
"The prior 10% discount and free premium support are treated as the baseline this renewal negotiates from, not ignored"
]
},
{
"id": 9,
"prompt": "Whitmore Aerospace, $460k/yr, four people involved: their CFO Dana, a procurement lead named Ruiz, their VP Engineering, and the COO who has been sponsoring this since January. Everybody wants something different - Dana is on payment terms, Ruiz wants a number he can report upward, engineering wants integration help and a higher uptime commitment, the COO wants it live before their busy season. I have some room on price but not a lot. How should I structure what I offer?",
"expected_output": "A ledger with a named best recipient per row, each concession routed to the stakeholder who values it most, re-ranked after routing, plus explicitly marked log-roll candidate pairs.",
"files": [],
"expectations": [
"Each concession is routed to the stakeholder who values it most, with a named best recipient recorded per ledger row",
"Payment terms and financial risk are routed to finance",
"Documented savings are routed to procurement",
"Speed and outcomes are routed to the business sponsor",
"Support, integration, and uptime levers are routed to the technical stakeholder",
"The output states that a lever's value is its value to that best recipient, not an average across the room",
"The ledger is re-ranked after the routing step rather than left in its pre-routing order",
"Explicit log-roll candidate pairs are marked: a seller low-priority issue against a buyer high-priority issue",
"The output states that multi-issue deals create value at exactly these trades and that negotiators miss them by default",
"It warns against giving procurement a concession the sponsor never asked for",
"Any reference-call get is capped in number rather than left open-ended"
]
},
{
"id": 10,
"prompt": "We shook hands Tuesday with Ledbrook Systems on $175k/yr. This morning their procurement guy emailed asking to add 10 seats at no extra cost, since we're basically done anyway. Worth knowing: the number we landed on already has our Q3 promo of 10%, a partner referral discount of 5%, and a 7% competitive match baked into it. Nothing is signed yet. How do I answer him?",
"expected_output": "The post-handshake ask named as a nibble and countered procedurally by reopening the whole package, the stacked discounts totalled before anything else is granted, and every trade written into the order form.",
"files": [],
"expectations": [
"The post-handshake ask is identified as a nibble",
"It is countered procedurally by reopening the entire package rather than granting the seats in isolation",
"The output states that a post-agreement ask reopens everything, which is what makes nibbles expensive for the asker",
"The conditional reopening sentence is written out verbatim for the seller to send or say",
"The combined concession across the Q3 promo, the partner referral discount, and the competitive match is computed before anything further is granted",
"The output flags that stacked promotions and discounts compound past what any single approval intended",
"Every granted trade and its reciprocal get is required to be written into the order form and the written recap",
"The output states that verbal-only concessions die with the champion",
"It recommends getting the buyer's full remaining list on the table before moving again, keeping all issues linked",
"The added seats are paired with a named reciprocal get and the approval level the combined concession now requires"
]
}
],
"trigger_queries": [
{ "query": "procurement came back asking for 22% off, how do I respond", "should_trigger": true },
{ "query": "build me a concession plan before my call tomorrow", "should_trigger": true },
{ "query": "what should I trade for a discount", "should_trigger": true },
{ "query": "the buyer wants 20% off", "should_trigger": true },
{ "query": "I need a give-get list for a contract conversation next week", "should_trigger": true },
{ "query": "how do I decide my walk-away point on this deal", "should_trigger": true },
{ "query": "what's my BATNA here and where exactly do I stop", "should_trigger": true },
{ "query": "they're asking for net-90 and a renewal cap, what do I give up", "should_trigger": true },
{ "query": "prep me for a redlines conversation with their legal team", "should_trigger": true },
{ "query": "customer wants me to match a competitor's price, what do I do", "should_trigger": true },
{ "query": "put together give-get trades for Friday's pricing call", "should_trigger": true },
{ "query": "we need three packages that cost us the same but look different to them", "should_trigger": true },
{ "query": "what can I offer instead of cutting the price", "should_trigger": true },
{ "query": "how do I get something back every single time I move on price", "should_trigger": true },
{ "query": "i'm going into a pricing conversation and I don't want to just cave", "should_trigger": true },
{ "query": "the CFO wants a three year price hold, how expensive is that for us", "should_trigger": true },
{ "query": "customer wants the monthly under $500, what levers do I have left", "should_trigger": true },
{ "query": "should I throw in free installation or just drop the price", "should_trigger": true },
{ "query": "make me a one pager for tomorrow's meeting with their purchasing team", "should_trigger": true },
{ "query": "how much am I allowed to give away and what do I ask for in return", "should_trigger": true },
{ "query": "they keep nibbling at me after we already shook hands", "should_trigger": true },
{ "query": "what do I do when they say you'll have to do better than that", "should_trigger": true },
{ "query": "buyer gave me a fixed budget number and told me to fit it", "should_trigger": true },
{ "query": "we're at the end of quarter and they know it, help", "should_trigger": true },
{ "query": "list everything I could trade in this contract that isn't price", "should_trigger": true },
{ "query": "plan the trade-offs for a renewal where they want an uplift cap", "should_trigger": true },
{ "query": "how do I structure this so it doesn't set a precedent for every renewal", "should_trigger": true },
{ "query": "i want to know what to concede first and what to hold back", "should_trigger": true },
{ "query": "the trade-in number they want is well above book value, what now", "should_trigger": true },
{ "query": "seller-side prep for a contract negotiation next week", "should_trigger": true },
{ "query": "what should I ask for in exchange for extended payment terms", "should_trigger": true },
{ "query": "help me plan what I'm actually willing to give away on this deal", "should_trigger": true },
{ "query": "procurement wants documented savings, how do I give them a win without losing margin", "should_trigger": true },
{ "query": "should we agree to most favoured customer pricing", "should_trigger": true },
{ "query": "i need to know when to walk away from this deal", "should_trigger": true },
{ "query": "what do I get back for adding two more modules at the same price", "should_trigger": true },
{ "query": "buyer wants termination for convenience, what do I counter with", "should_trigger": true },
{ "query": "structure the discount so it doesn't blow up our renewals", "should_trigger": true },
{ "query": "they want a pilot at half price, is that a good trade for us", "should_trigger": true },
{ "query": "prep for a purchasing department call, they're going to squeeze me", "should_trigger": true },
{ "query": "what's the cheapest thing I can hand over that they'll value most", "should_trigger": true },
{ "query": "i keep discounting and getting nothing back for it", "should_trigger": true },
{ "query": "negotiation prep for a $200k software deal closing this month", "should_trigger": true },
{ "query": "how do I respond when they claim they can't approve that themselves", "should_trigger": true },
{ "query": "map out what each side wants and where we could swap", "should_trigger": true },
{ "query": "i want to stop bidding against myself on price", "should_trigger": true },
{ "query": "customer says another dealer quoted him lower but won't show me anything", "should_trigger": true },
{ "query": "help me plan the concessions for an enterprise renewal", "should_trigger": true },
{ "query": "what approvals do I need for each thing I might give away", "should_trigger": true },
{ "query": "write the if-then lines I should use when they push on price", "should_trigger": true },
{ "query": "the prospect says we're too expensive, what should I say back", "should_trigger": false },
{ "query": "write me a rebuttal for when they say they already use a competitor", "should_trigger": false },
{ "query": "build an objection library for my SDR team", "should_trigger": false },
{ "query": "what do I say when they tell me to just send some info", "should_trigger": false },
{ "query": "how do I handle a status quo objection on a stalled deal", "should_trigger": false },
{ "query": "build the ROI business case for this deal", "should_trigger": false },
{ "query": "what's the payback period if they buy 200 seats", "should_trigger": false },
{ "query": "calculate TCO against the incumbent for their CFO", "should_trigger": false },
{ "query": "quantify the cost of doing nothing for this prospect", "should_trigger": false },
{ "query": "score this opportunity against MEDDPICC", "should_trigger": false },
{ "query": "run BANT on this inbound lead", "should_trigger": false },
{ "query": "is this deal actually real or am I wasting my time", "should_trigger": false },
{ "query": "map the buying committee for this account", "should_trigger": false },
{ "query": "we're single threaded, who else should I be talking to", "should_trigger": false },
{ "query": "review my call notes and flag the risks in this deal", "should_trigger": false },
{ "query": "this deal went quiet three weeks ago, what now", "should_trigger": false },
{ "query": "design the accelerator curve for our AE comp plan", "should_trigger": false },
{ "query": "should we cap commissions above 200% attainment", "should_trigger": false },
{ "query": "how much quota should an enterprise AE carry next year", "should_trigger": false },
{ "query": "we had 4x coverage and still missed, what's wrong", "should_trigger": false },
{ "query": "should we go product-led or stay sales-led", "should_trigger": false },
{ "query": "define our ideal customer profile from last year's closed won deals", "should_trigger": false },
{ "query": "everything drifted into tier 1, fix our account tiers", "should_trigger": false },
{ "query": "build an account fit score from twelve months of deal data", "should_trigger": false },
{ "query": "estimate the TAM for mid-market logistics software", "should_trigger": false },
{ "query": "what do I say in the first ten seconds when they pick up", "should_trigger": false },
{ "query": "nobody opens my emails, rewrite these subject lines", "should_trigger": false },
{ "query": "why are my cold emails landing in spam", "should_trigger": false },
{ "query": "how many touches should my outbound cadence have", "should_trigger": false },
{ "query": "find me a reason to reach out to this prospect this week", "should_trigger": false },
{ "query": "what discovery questions should I ask on a first call", "should_trigger": false },
{ "query": "write the follow up email after yesterday's meeting", "should_trigger": false },
{ "query": "score this call transcript and tell me what I did wrong", "should_trigger": false },
{ "query": "interview questions for hiring an SDR", "should_trigger": false },
{ "query": "how do I break into tech sales with no experience", "should_trigger": false },
{ "query": "pods or assembly line for a twelve person sales team", "should_trigger": false },
{ "query": "which sales podcasts and newsletters should I be following", "should_trigger": false },
{ "query": "which sales skill do I need for this project", "should_trigger": false },
{ "query": "negotiate my salary for this offer letter", "should_trigger": false },
{ "query": "how do I ask my manager for a raise", "should_trigger": false },
{ "query": "counter this job offer, the base is too low", "should_trigger": false },
{ "query": "review this MSA and flag the risky clauses for us", "should_trigger": false },
{ "query": "what's wrong with the indemnity language in this vendor agreement", "should_trigger": false },
{ "query": "should we move from seat-based to usage-based pricing", "should_trigger": false },
{ "query": "set list prices for our new enterprise tier", "should_trigger": false },
{ "query": "design the three tiers on our pricing page", "should_trigger": false },
{ "query": "i'm buying a used car this weekend, how do I get them down on price", "should_trigger": false },
{ "query": "we're the buyer here - how do I push this vendor to come down 20%", "should_trigger": false },
{ "query": "negotiate our electricity supply contract for the next three years", "should_trigger": false },
{ "query": "help me negotiate a lower rent with my landlord", "should_trigger": false }
]
}
references/concession-inventory.md›
# Concession Inventory - Ranked Tradeable Levers
Catalogue for building the bargaining mix (workflow step 1) and filling the ledger (step 3). Every lever arrives priced and ordered, so the plan inherits a ranking instead of re-deriving one per deal.
**Cost to us** is never a currency figure. It is margin exposure, the precedent the concession sets for every deal after this one, the delivery burden it creates, and how hard it is to take back - scored in magnitudes only: near-zero, an hour, a week, a quarter, a standing job.
"A standing job" means it never ends: a lever you can never use again, a clause that binds every renewal. A concession that becomes the new baseline is expensive whatever its face value looks like today.
**Value to them** is the outcome it buys: a sign-off unblocked, a budget line that fits, a number procurement can report upward.
**Compliance cost** is named only on levers carrying contractual or regulatory exposure - the review they trigger and how long you are stuck with them. Most levers carry none, and those rows say nothing about it.
Rank by value to them per unit of cost to us, and lead with the best ratio, not with the cheapest lever. Every order below is a default, not a law: it shifts with the deal and with who executes it. Re-rank before the ledger:
- A renewal already carries last time's concessions in its baseline, so price protection means something different there than on a new logo.
- A buyer who has already anchored on a number leaves you only the non-price rows.
- A rep who needs deal-desk sign-off for a whole class of lever should treat that class as one tier lower than it reads here.
## Table of Contents
- [B2B gives - ranked](#b2b-gives-ranked)
- [B2B gets - ranked](#b2b-gets-ranked)
- [B2C gives - ranked](#b2c-gives-ranked)
- [B2C gets - ranked](#b2c-gets-ranked)
## B2B gives - ranked
The axes disagree sharply, so each gets its own line (families, ordered):
- **efficiency (act on this):** service access > cash timing > scope > headline discount > price protection
- **cost to us:** price protection > headline discount > scope > cash timing == service access
- **value to them:** price protection > headline discount > scope > cash timing > service access
The `==` holds because cash timing and service access cost the same kind of thing: capacity or working capital you get back, inside the term, with no clause left behind. Neither sets a precedent the next renewal inherits - which is what separates both of them from everything to their left.
| Lever | Family | Cost to us | Value to them | Tier |
| --------------------------------------------------------------------------- | ---------------- | ----------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
| Sandbox / non-production environment | service access | near-zero - the infrastructure already runs, and it comes back at renewal | unblocks IT and security, usually the last gate before signature | first move |
| Named CS contact and escalation path | service access | an hour a week, fully reversible | the sponsor's confidence that rollout will not stall | first move |
| Data export window, security addendum, modest liability cap | risk | an hour of legal review each, on standard language | clears the security review that otherwise runs a month | first move - compliance: legal sign-off per deal, bounded to the term |
| Delayed start date | cash timing | near-zero - revenue slips, it does not vanish | fits a budget year that is already spent | first move |
| Delayed first invoice | cash timing | near-zero - the same mechanism, moving the invoice clock instead of the service clock | same | first move |
| Extended payment terms (net-45/net-60) | cash timing | near-zero when your cash position allows it, a quarter of working capital when it does not | finance's own scoreboard; routinely beats a discount of equal perceived size | first move |
| Uptime commitment upgrade and service credits | risk | near-zero when your reliability already clears the bar, a standing job when it does not | high during an incident, higher inside procurement's risk review | first move - compliance: legal and ops sign-off; credits stay a live liability for the whole term |
| Overage forgiveness on the first true-up; annual instead of monthly true-up | scope | an hour of ops, and a soft precedent the next true-up inherits | removes the fear of a surprise bill nobody budgeted | first move |
| Extra training days, implementation or migration hours | service access | a week of delivery capacity, one-off | time-to-value the sponsor is personally measured on | first move |
| Premium support tier, time-boxed to year 1 | service access | a quarter of support capacity, reversible only if the time box is in writing | high through rollout, close to irrelevant after it | mid |
| Ramped pricing, lower year 1 stepping up | commercial | a quarter of year-1 revenue, recovered later only if the ramp actually holds | a budget that grows with adoption instead of preceding it | mid |
| Pilot pricing on a new module, time-boxed | commercial | near-zero when the box genuinely closes, a standing job when the pilot quietly renews at pilot price | lets them buy before they can build a business case | mid |
| Multi-year term with an annual opt-out | commercial | the opt-out is the cost, not the term - it turns a commitment into a rolling deal; price the two separately | commitment without the risk of being wrong | mid |
| Extra seats or usage capacity | scope | margin now, and the larger number becomes the baseline they renew on | headline capacity at no headline price | mid |
| Extra modules bundled at the same price | scope | as above, plus the bundle becomes their reference price forever | the breadth their business case needs | mid |
| Flexibility in how billable usage is defined | scope | an hour to draft, a standing job to police, and hard to unwind | predictability finance can actually forecast | mid - compliance: deal desk and revenue accounting; the definition follows every renewal |
| Headline discount | price | direct margin this year, and the number every future renewal negotiates down from | the win procurement reports upward | mid, sliding to last resort with depth |
| Renewal uplift cap | price protection | a standing job - your main net-revenue-retention lever, gone | often worth more to them than another point off year 1 | last resort - compliance: deal desk and finance; irreversible across the capped horizon |
| Price hold across renewal cycles | price protection | a standing job - the same lever removed, differing only in degree | same | last resort - compliance: as above |
| Most-favoured-customer treatment | price protection | a standing job across every deal you sign after this one | insurance against being the customer who paid more | last resort - compliance: legal and finance; keep it narrow (list pricing only, prospective, no audit rights) or refuse |
| Termination for convenience | risk | undoes any term-based discount and can block revenue recognition | an exit that needs no business case | last resort - compliance: revenue accounting and legal; counter with locked initial-term pricing or full-term prepay |
| Liability super-cap, broad indemnity | risk | a standing job an acquirer's diligence reprices years later | their risk committee's checkbox | last resort - compliance: legal and executive sign-off; survives the contract itself |
Ties above: delayed start `==` delayed first invoice - one mechanism, two clocks, identical cost and identical value. Renewal uplift cap `==` price hold - both remove next renewal's price as a lever, and differ only in how completely.
**What this order starves:** price protection. It sits last on efficiency precisely because it is the most expensive family on the page, so a ratio-led plan never reaches it - and it is exactly what a buyer whose real constraint is year 2 and year 3 came for.
Promote it above the headline discount when their pressure is about renewal predictability rather than this year's price. A capped uplift costs nothing in the current period, buys more deal than a discount of the same felt size, and pairs naturally with the term length that pays for it.
Service and delivery-heavy levers get starved from the other direction in procurement-led deals, where nobody on the buying side is scored on rollout succeeding - promote them when the sponsor, not procurement, holds the pen.
## B2B gets - ranked
For gets the ratio inverts: value to us per unit of friction for them. Ask high-ratio gets against small gives; save the bottom of the list for the levers that actually hurt.
- **efficiency:** signature date > clause removal > introductions > prepay and billing > logo > bounded references > case study > multi-year > expanded scope
- **value to us:** multi-year == expanded scope > prepay and billing > case study > clause removal > bounded references > logo > introductions > signature date
The `==` is real: a multi-year commitment and a scope expansion both convert a promise into booked revenue, which is the only currency on this list that shows up in the same place a discount does.
| Get | Cost to them | Value to us | Ask it against |
| ------------------------------------------------------------------- | -------------------------------------------------------------- | ---------------------------------------------------------------- | --------------------------------- |
| A committed signature date | near-zero - a date is not a budget | the forecast closes; the only get that costs them nothing at all | every trade, however small |
| Removal of a buyer-side clause they requested (TFC, MFN, super-cap) | near-zero whenever they did not truly need it | takes a standing job back off your side of the contract | any price-protection or risk give |
| Referrals or introductions to named peers | near-zero, ten minutes of their time | pipeline with no acquisition cost attached | first-move gives |
| Annual prepay, or annual instead of quarterly billing | a quarter of their cash flexibility | cash now, and churn insurance for the year | cash-timing gives |
| Public logo rights | an hour of their marketing and legal | proof in every deal you run after this one | first-move and mid gives |
| Reference calls, capped in number | an hour per call - a standing job if you leave the number open | evidence in every competitive cycle | mid gives |
| A published case study | a week of their comms and legal | the strongest evidence asset there is | mid gives |
| Multi-year commitment | a quarter of their budget flexibility | the term that funds a ramp or a cap | last-resort gives |
| Expanded scope or seats | real budget they have to go and find | revenue rather than a promise | last-resort gives |
Never leave reference calls uncapped - an open commitment is the one get on this list whose cost to them grows without limit, and they will price it that way the moment they notice.
## B2C gives - ranked
- **efficiency:** slack capacity > perceived-value goods > payment structure > delivery labour > headline price
- **cost to us:** headline price > delivery labour > payment structure > perceived-value goods > slack capacity
- **value to them:** headline price > payment structure > delivery labour > perceived-value goods > slack capacity
| Lever | Family | Cost to us | Value to them | Tier |
| ----------------------------------------------------- | ------------------ | ------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| Old-unit removal, haul-away | retail | near-zero when the truck is already going that way | a chore they were quietly dreading, gone | first move |
| Priority scheduling, a slot that fills slack capacity | retail | near-zero, sometimes better than zero - it fills a gap you were carrying | decisive whenever they have a date of their own | first move |
| Closing-date flexibility | real estate | near-zero when nothing is chained behind you | frequently the whole deal for a buyer moving between homes | first move |
| Gift card instead of a headline discount | retail | under face value once breakage and margin are counted, and it leaves your reference price intact | reads as cash off the price | first move |
| Free or upgraded delivery | retail | an hour of route capacity | perceived at several times what it costs you | first move |
| Bundled accessories or consumables | retail | unit cost, one-off | completes a purchase they had already decided on | first move |
| Down payment and monthly payment restructure | automotive | near-zero to move value between price, trade, down payment and monthly - this is the core B2C log-roll | the monthly number is the one they are actually buying | mid |
| Installation and setup | retail | a week of installer capacity in season, near-zero out of it | removes the reason they postpone the purchase | mid |
| Included appliances, home warranty | real estate | unit cost, one-off | a house they can move into without a shopping list | mid |
| Trade-in allowance above book value | automotive | margin, offset elsewhere in the deal, and reversible deal by deal | the number they will repeat to everyone they know | mid |
| Extended warranty | retail, automotive | the reserve behind it, not its sticker | peace of mind on a purchase they cannot easily repeat | mid - compliance: warranty and disclosure rules vary by jurisdiction; sell it, never gift it |
| Repairs after inspection | real estate | open-ended until the punch list closes - a standing job you signed for blind | their inspection risk transferred to you | last resort - compliance: disclosure obligations attach to whatever you agree to fix |
| Financing terms (deferred payment, 0% period) | automotive, retail | the buy-down, plus the obligations that come with arranging credit at all | turns an unaffordable monthly into an affordable one | last resort - compliance: consumer-credit disclosure, regulated in most jurisdictions; reversible only by unwinding the sale |
| Closing-cost credit | real estate | direct margin, visible to everyone on the settlement statement | cash at the moment they have the least of it | last resort - compliance: lender limits on seller credits |
| Headline price cut, or a match to a written quote | all | direct margin, and your own reference price with it | their stated target, exactly | last resort - compliance: advertised-price rules wherever the cut gets published |
**Deleted, not demoted: the price-match guarantee.** A standing promise to match any future competitor price is ruled out by this skill's own operating rule - answer a written quote by pricing the scope difference, never by matching a number. A guarantee cannot be paired with a get, cannot be capped, and re-opens the price after the sale has closed. It is off the list rather than parked at the bottom, because a lever parked at the bottom comes back into the room.
**What this order starves:** payment structure and financing. Restructuring the monthly is high-value and, on the financing end, genuinely expensive and regulated - so a ratio-led plan reaches for goods and slack capacity and never gets there. Promote payment structure to a first move the moment the buyer's objection is the monthly rather than the price; those are different objections and only one of them is about your margin.
## B2C gets - ranked
- **efficiency:** purchase today > review and referral commitment > pickup or flexible scheduling > back-end product attachment
- **value to us:** back-end attachment > purchase today > review and referral > pickup or scheduling
Back-end products (service contracts, protection plans) are frequently where the real margin lives, which is why they sit last on efficiency and first on value: they cost the buyer actual money. A front-end price concession funded by back-end attachment is one package - price it as one, and never grant the front end before the attachment is agreed.
references/concession-ledger-example.md›
# Concession Ledger - Filled Examples
Illustrative numbers, invented for format demonstration - replace with the actual deal's figures.
## B2B SaaS example
Deal: workflow platform, $120,000/yr list (100 seats), enterprise buyer.
- Procurement has asked for 20% off.
- Sponsor wants a go-live before their fiscal year starts.
- Seller's quarter ends in 5 weeks (noted as a risk, not a lever).
```
WALK-AWAY: $96,000/yr at current scope - below this, redeploying the AE's time
to two mid-funnel deals of similar size is worth more than this contract.
DEAL SPACE: their alternative (incumbent renewal + manual work) estimated at
~$105k equivalent cost. ZOPA exists roughly $96k-$110k at current scope.
```
Rows sit in ratio order inside each tier - best value per unit of cost first, not cheapest first. The sandbox leads the plan over the training days even though both are cheap: one costs infrastructure that already runs, the other costs a week of delivery capacity that is not free just because it is already paid for.
| Lever | Tier | Cost to us | Value to them | Best recipient | Required get | Approval | Range (open → limit) |
| -------------------------- | ----------- | ------------------------------------------------------------------------- | --------------------------------------------------------------------------- | -------------- | --------------------------------------- | ------------------------------------------------------------------- | -------------------- |
| Sandbox environment | Cheap | near-zero - the infrastructure runs already, and it comes back at renewal | unblocks IT sign-off, the last gate before signature | Technical eval | Reference call commitment (2/yr) | AE | included |
| Net-45 payment terms | Cheap | near-zero at the current cash position | finance stops opposing the deal | Finance | Annual (not quarterly) billing | AE | net-45 → net-60 |
| Extra training days (2) | Cheap | a week of delivery capacity, one-off | go-live lands before their fiscal year opens | Technical eval | Signature date this month | AE | 1 day → 3 days |
| Premium support tier, yr 1 | Mid | a quarter of support capacity, reversible because the time box is written | covers the rollout risk the sponsor is personally measured on | Sponsor | Case study after go-live | Manager | 6 mo → 12 mo |
| Ramped pricing (80% yr 1) | Mid | a quarter of yr-1 revenue, recovered only if the ramp holds | fits a budget that grows with adoption instead of preceding it | Finance | 3-year term, yr 2-3 at full price | Manager | 90% → 80% yr 1 |
| 8% discount | Mid | direct margin, and the baseline every renewal starts from | the number procurement reports upward | Procurement | 2-year term | Manager | 5% → 10% |
| 15% discount | Last resort | the same, deeper, and it resets the reference price permanently | their stated target | Procurement | 3-year term + full prepay + logo rights | Deal desk + finance | 12% → 15% |
| Renewal uplift cap 5%/yr | Last resort | a standing job - the net-revenue-retention lever, gone | renewal-year predictability, worth more to them than another point off yr 1 | Procurement | Drop the TFC clause they requested | Deal desk + finance - binds every renewal across the capped horizon | 7% → 5% |
MESO packages - `A == B == C` on cost to the seller by construction; if one were cheaper for us than the others it would be a discount ladder, not a MESO:
- **A - Price-led**: $108k/yr, 2-year term, standard support. For a procurement-driven close.
- **B - Cash-led**: $114k/yr, 1-year term, full prepay, net-30, premium support 6 months. For a finance-driven close.
- **C - Term-led**: $102k/yr average via ramp (80/105/110%), 3-year term, uplift capped at 5%. For a budget-cycle-driven close.
Log-roll candidates:
- Our payment-terms flexibility (low cost to us) ↔ their term-length flexibility (low cost to them, high value to us).
- Our sandbox/training (near-free) ↔ their references and case study (near-free to them).
Hardest asks to rehearse:
- "If we cap renewals at 5%, we need the termination-for-convenience clause removed. Those two travel together."
- "A 15% discount is a board-level price for us - it only exists with a 3-year prepaid commitment. Is that on the table for you?"
## B2C example
Deal: appliance retailer, $2,400 washer-dryer package. Customer asked "what can you do on price?" and mentioned a competitor's $2,250 online listing.
| Lever | Tier | Cost to us | Value to them | Required get | Approval | Range |
| --------------------- | ----------- | --------------------------------------------------------------------------------------- | --------------------------------------------- | ----------------------------------- | ----------- | --------------------------- |
| Old-unit removal | Cheap | near-zero - the truck is already going to the address | a chore they were dreading, gone | Purchase today | Salesperson | included |
| Free delivery | Cheap | an hour of route capacity | reads as worth several times what it costs us | Purchase today | Salesperson | included |
| $100 gift card | Mid | under face value once breakage and margin are counted, and the reference price survives | reads as cash off the price | Review + referral commitment | Manager | $50 → $100 |
| Installation | Mid | a week of installer capacity, in season | removes the reason they postpone the purchase | Extended-warranty attachment | Salesperson | half price → free |
| Price match to $2,250 | Last resort | direct margin, and the reference price with it | their stated target, exactly | Warranty + install package purchase | Manager | match only vs written quote |
Note the structure: rows in ratio order, so the headline price moves last and only against a written competitor quote, while slack capacity and the perceived-value goods absorb the pressure first. Cost cells carry magnitudes, never the item's invoice cost - the gift card's face value is a deal figure, its cost to us is the class of thing it spends.
## Negative example - the unconditional give
> Buyer: "Your competitor is cheaper."
> Seller: "I hear you - let me take 15% off if that helps get this done this week."
Three defects:
- It is a give without a get (the buyer learns the next ask is also free).
- It concedes before the buyer specified what "cheaper" means or showed the quote.
- It burns the mid-tier and part of the last-resort tier in one move, leaving nothing for the real negotiation.
**Corrected:**
> Seller: "Can you share that quote? If the scope really matches ours, then - with a signature by the 28th and the two-year term - I can bring us to $X. Without those, $Y is where this package sits."
Same commercial flexibility, but now conditioned, evidence-gated, paired with two gets, and inside the planned range.
references/trade-language.md›
# Trade Language - Conditional Framing Lines
Every planned trade gets its sentence written before the call. The template is always conditional: **"If you can do X, then we can consider Y."** The get comes first in the sentence - it anchors the trade as an exchange, not a gift.
## Lines by buyer ask
Keyed to what the buyer said, not ranked - you do not choose between these, the buyer does.
**Discount ask**
- "If you can commit to a two-year term, then we can look at the price."
- "Help me understand what number you need to get to, and what you're able to bring to the table to get there."
**Payment-terms ask (net-60, quarterly billing)**
- "If we move to annual billing, then net-60 works on our side."
- "Extended terms are doable - in exchange we'd need the start date locked to this month."
**Price-hold / renewal-cap ask**
- "A renewal cap is one of the most expensive things we can grant. If it matters that much, then we'd pair it with a three-year commitment."
**Extra-scope ask (more seats, extra module, added service)**
- "We can include the sandbox environment - and in return we'd ask for two reference calls a year once you're live."
- "If we add the training days, then we'd want the case study commitment in the order form."
**The vague "you'll have to do better than that"**
- "Happy to look at it - what specifically needs to improve, and by how much?" (Then trade against the specific answer; never move on the vague version.)
**Budget bogey ("we only have $X")**
- "If $X is truly fixed, then let's fit the scope to $X - here's what the package looks like at that number." (Reduce scope, not price.)
**Competitor price claim**
- "If you can share that quote in writing and the scope matches, then we'll respond to it seriously. Where the scope differs, I'll price the difference for you."
**The last-minute nibble (small ask after handshake)**
- "We can add that - it reopens the package, so let's look at the whole agreement together." (A post-agreement ask reopens everything; that is what makes nibbles expensive for the asker.)
**Signaling the limit (decelerating pattern, final step)**
- "That's the end of our range. I can add [small non-price item], but the number itself is done."
## Lines for making the asks (the gets)
Ranked by value to you per unit of friction for them, matching the ranked gets list in the concession inventory: a signature date first, expanded scope last. Ask down the list, and stop as soon as the give you are pairing it with is covered.
- "Can we agree on a signature date now? The package I've shaped assumes we close this month." (Costs them nothing at all - pair it with any give, however small.)
- "If we drop the termination-for-convenience clause, then the term pricing works. Those two travel together." (Near-zero for them whenever they did not truly need the clause.)
- "Who else in your network faces this problem? Two introductions would mean a lot to us - and cost you ten minutes."
- "If prepay is impossible, is a multi-year term possible instead? One of the two makes this price work."
- "For a partnership at this price level, we'd want you as a referenceable customer - two calls a quarter, capped." (Never ask for references uncapped; an open commitment is the one get whose cost to them grows without limit.)
## What NOT to say
- ~~"Let me see what I can do."~~ - Concedes that movement exists before any trade is named. Say instead: "What would you be able to do on your side?"
- ~~"Our list price is X, but..."~~ - Volunteering that the list price is soft invites the discount conversation before the buyer even pushed. State one price; let them push first.
- ~~"I can do 10% off if that helps."~~ - Unconditional give. Every number travels with its condition.
- ~~"I really need this to close this quarter."~~ - Hands the buyer your calendar as leverage. Your timing pressure never enters the room.
- ~~"It's a fair price because of A, and also B, and also C, and honestly D too."~~ - Argument dilution: the buyer attacks the weakest reason and the whole position wobbles. One strong reason, then silence.
- ~~"Fine, we'll match their price."~~ (against a verbal claim) - Matching an unverified number rewards bluffing. Written quote first, scope comparison second, response third.
SKILL.md›
---
name: negotiation-concession-planner
description: Builds a pre-negotiation concession plan for a pricing or contract conversation - every tradeable lever priced by cost-to-us vs value-to-them, tiered, each give paired with a required reciprocal get and its approval level, plus a BATNA-based walk-away. Covers B2B (terms, scope, service, risk, price protection) and B2C (bundles, financing, trade allowance). Use whenever the user mentions a discount request, procurement, redlines, give-get trades, walk-away point, or "the buyer wants 20% off", even without the word negotiation. Do NOT use for quantifying the value (mbfinotti/sales-skills@deal-value-calc) or scripting price rebuttals (mbfinotti/sales-skills@sales-objection-handling).
license: MIT
metadata:
author: Maya-Beth Finotti
version: "1.4.8"
---
# Concession Planner
Produce a pre-call concession plan: a tiered ledger of what the seller can trade, priced by cost-to-us vs value-to-them, every give paired with a required reciprocal get and the approval level it needs, bounded by a written walk-away line.
A concession plan is a preparation artifact, not an in-call improvisation. Mid-negotiation, emotional engagement degrades judgment of relative value - so the gives, the gets, the prices, and the limits get written before the conversation starts.
## Interview
Ask before planning. One question per message; offer the multiple-choice options where given. Skip anything already answered by prior context.
1. What are you selling, and at what price point is the deal on the table?
2. Is this B2B or B2C, and which segment: (a) enterprise, (b) SMB / mid-market, (c) consumer?
3. Who is the counterpart: (a) the decision maker, (b) procurement / purchasing, (c) a champion or influencer without final authority, (d) unknown?
4. What has the buyer actually asked for so far: (a) a price discount, (b) better terms - payment, contract length, risk, (c) more scope for the same price, (d) nothing yet, but you expect pushback?
5. What is your discount authority band, and who approves beyond it? Treat the existing approval matrix as an input - this skill never designs it.
6. Which non-price levers can you move: term length, payment timing, scope or volume, service level, risk terms (uptime, credits, liability), price protection, bundled goods?
7. What do you want in return: (a) references, case study, or logo rights, (b) prepay or multi-year, (c) a faster signature date, (d) referrals or introductions, (e) larger scope?
8. What is your walk-away alternative if no deal happens - the next-best use of the same time, capacity, or inventory?
9. By what date must this land, and whose date is it: (a) theirs, (b) yours, including your own quarter-end, (c) neither's?
10. Do you want a one-off win here, or a position that holds? A concession granted once tends to be expected forever - every renewal starts from it, and your other buyers hear about it. (a) close this one at almost any structure, (b) protect the baseline for the renewal and the deals after it.
11. What is your effort ceiling: delivery capacity you can actually commit, approvals you can realistically get in the time available (deal desk, finance, legal), and how much of this you can take back later?
Re-rank the inventory against those three answers before building the ledger, and say which answer moved what:
- A hard, near date promotes the fast-acting levers - cash timing, service access, a delivery commitment - and demotes multi-year structures that need approvals you cannot collect in time.
- A "protect the baseline" answer demotes the headline discount and every price-protection lever, since both are permanent by construction. "Close it now" is the only answer that promotes them above their tier.
- A low effort ceiling **deletes** levers rather than demoting them: anything needing legal or deal-desk sign-off you cannot get, anything creating delivery work you cannot staff. Strike them from the ledger and name which ones you struck - a ruled-out lever parked at the bottom reappears in the room as an offer.
If the counterpart cannot decide (3b without a mandate, or 3c/3d), flag it before planning: a concession granted to a non-decision-maker gets spent twice, because the real decision maker will ask for it again.
## Workflow
1. **Build the bargaining mix.** List every issue in play - price, terms, scope, service, risk, timing - including the issues the buyer will raise, not only the ones you plan to. Lay out the four levers (volume, cash timing, contract length, deal timing) before touching price, so you can trade across them instead of defaulting to a discount. Pull levers from [concession-inventory.md](./references/concession-inventory.md), already priced and ranked, minus the ones question 11 ruled out.
2. **Write two separate lists**: what you are willing to give, and what you will demand in exchange. Write both now, in full - in the room you will judge relative value poorly and forget half your asks.
3. **Price every item on cost-to-us vs value-to-them, and order it by the ratio.** Lead with the highest value per unit of cost, not with the cheapest lever.
- Cost to us is margin, the precedent set for every deal after this one, delivery burden, and reversibility - never a currency figure, scored only in magnitudes: near-zero, an hour, a week, a quarter, a standing job.
- Default B2B family order, efficiency axis: `service access > cash timing > scope > headline discount > price protection`.
- Cost-to-us axis: `price protection > headline discount > scope > cash timing == service access` (the tie holds because both spend capacity or working capital you get back inside the term, leaving no clause behind).
- Value-to-them axis: `price protection > headline discount > scope > cash timing > service access`.
- [concession-inventory.md](./references/concession-inventory.md) carries every lever priced against those axes, the B2C order, and what each order starves.
- Treat that order as a default, not a law: re-rank it against this deal and against who has to execute it. A renewal negotiates up from last time's concessions, a buyer already anchored on a number leaves you only the non-price rows, and a lever class your rep cannot get approved in time drops a tier.
4. **Tier the ledger in ratio order**: first moves, mid, last resort, highest value per unit of cost at the top of each tier, never the cheapest item first.
- The same concession is worth different amounts to different people:
- Finance weighs payment terms and financial risk.
- Procurement weighs documented savings.
- The business sponsor weighs speed and outcomes.
- Technical roles weigh support and integration.
- Route each concession to the stakeholder who values it most, then re-rank. A lever's value is its value to that best recipient, not an average across the room.
- Pair every item with (a) a named reciprocal get and (b) the approval level it needs. For a lever carrying contractual or regulatory exposure, record its compliance cost and how long it binds you in that approval cell.
- No orphan gives: an unpaired concession is a defect in the plan, not a nice gesture.
5. **Set the walk-away from your alternative, never from your margin floor.** Value the no-deal alternative (your BATNA), derive the reservation price from it, and write it down. Estimate the counterpart's reservation point too: if the two ranges cannot overlap (no ZOPA), walking is the rational plan and no concession sequence fixes it.
6. **Plan ranges per issue and plan issues independently, not as a fixed sequence.** Set an opening, a target, and a limit for each issue.
Do not script the order you will concede in: field observation of skilled vs average negotiators (Rackham & Carlisle) found heavy sequence planning is what average negotiators do. The order is itself negotiable, and a pre-set order collapses the moment the counterpart opens on your last item.
Plan each issue so you can trade in whatever order the conversation takes. The ratio ranking from step 3 says which lever to reach for first when you do move; it is not a script for the order the issues come up in.
7. **Build 2-3 MESO packages** (multiple equivalent simultaneous offers): packages of equal value to you but different shape for the buyer - e.g., lower price with shorter term vs higher price with longer term and stronger support. Do not rank them: `A == B == C` on cost to us by construction, and if one is genuinely cheaper for you than the others it is not a MESO, it is a discount ladder wearing three hats. Which one they gravitate to reveals their priority ordering without you asking for it.
8. **Plan the log-rolls.** Where your priority ranking differs from theirs, trade your low-priority issues for their high-priority ones. Multi-issue deals create value exactly here, and negotiators miss these trades by default - mark the candidate pairs in the plan explicitly.
9. **Shape a decelerating pattern within each issue**: shrinking increments toward your limit signal that the limit is real. Honest caveat: the supporting evidence is lab-based and mostly single-issue; in a multi-issue deal, log-rolling (step 8) matters more than taper mechanics. Hold your pre-set target no matter what pattern the other side runs at you.
10. **Write the conditional line for every planned trade**: "If you can do X, then we can consider Y." Never plan an unconditional give. Draft the exact sentences now from [trade-language.md](./references/trade-language.md) - improvised phrasing under pressure turns trades into gifts.
11. **Produce the pre-call one-pager** (format below) and rehearse the two or three hardest asks out loud - the get requests, not the gives.
12. **Run the Measurement check** below; iterate on the plan until it passes.
If your harness has persistent memory, store the walk-away line, authority band, and remaining unspent tiers after each round - the next conversation on the same deal starts from the updated plan, not from scratch. Without memory, tell the user to keep the one-pager and mark spent concessions on it.
## B2B vs B2C
Every workflow step above applies identically to both - only the concession currency differs.
- B2B currency is contract terms whose cost arrives later: term length, payment timing, ramped pricing, scope, support tier, uptime and service credits, renewal-uplift caps, price holds, liability caps. That delay is why sellers systematically under-price them.
- B2C currency is bundled goods with a known unit cost: delivery, installation, extended warranty, trade allowance, financing terms, priority scheduling, accessories, a gift card instead of a headline discount (protects the reference price). Cheaper to price accurately, easier to give away thoughtlessly.
- B2C compresses the timeline: single session, counterpart in the room. Keep the plan to one page and memorize the tiers - there is no pause to consult a ledger.
Both catalogues, ranked and priced: [concession-inventory.md](./references/concession-inventory.md).
## Pre-call one-pager
Output the plan in this shape (filled examples in [concession-ledger-example.md](./references/concession-ledger-example.md)):
```
DEAL: <what, list price, counterpart, their role, timing pressure each side>
WALK-AWAY: <reservation point + the alternative it derives from>
DEAL SPACE: <estimate of their reservation point; ZOPA yes/no>
TRADE LEDGER (rows in ratio order within each tier - best value per unit of cost first)
| Lever | Tier | Cost to us | Value to them | Best recipient | Required get | Approval | Range (open → limit) |
STRUCK LEVERS: <ruled out by the effort ceiling or authority - listed once, never carried in the ledger>
MESO PACKAGES: A == B == C on cost to us, different shape
LOG-ROLL CANDIDATES: <our low-priority issue ↔ their high-priority issue>
HARDEST ASKS TO REHEARSE: <2-3 get requests, verbatim>
```
Write "Cost to us" in magnitudes only - near-zero, an hour, a week, a quarter, a standing job - never as a currency figure. The deal's own prices belong on the DEAL and WALK-AWAY lines, not in the axis cells.
## Failure modes
These are not ranked, and deliberately so: they are not alternatives you pick between. Several can fire in the same call, and which one arrives first is the buyer's choice rather than yours - an order here would read as a queue to work through and leave the rest unguarded.
- **Unilateral discounting** - a give without a get teaches the buyer the next ask is also free. Fix: conditional line on every trade, no exceptions.
- **Conceding before pushback** - moving before the buyer has actually pressed is bidding against yourself. Hold the plan until an ask lands.
- **The sucker pattern** - one big opening concession then nothing signals the open was padded. Open small; keep increments shrinking.
- **Salami slicing and the last-minute nibble** - many small asks, or one "tiny" ask after handshake. Counter procedurally: get their full list on the table before moving, keep all issues linked, and re-open the whole package if a new ask appears after agreement.
- **Conceding to the wrong stakeholder** - giving procurement what the sponsor never asked for converts a relationship asset into a commodity input. Route per step 3.
- **Negotiating with a non-decision-maker** - every concession will be re-requested by the person who actually decides. Confirm authority before spending tiers.
- **Quarter-end capitulation** - the seller's own commission timing drives the deepest discounts, and buyers schedule to it. Decouple the plan from your fiscal calendar; cap what quarter-end is allowed to unlock.
- **Discount stacking** - promotions, bundle incentives, competitive matches, and discretionary discount compound past what any single approval intended. Compute the combined concession before granting the last piece.
- **Verbal-only concessions** - anything not written on the order form dies with the champion. Every granted trade and its get goes into the written recap and the order form.
- **Argument dilution** - extra weak reasons hand the counterpart something to attack. One strong reason per position; stop talking.
## Procurement counter-tactics
Procurement is measured on reportable savings against a baseline - they need a documented win more than a genuinely cheaper deal. A credible anchor plus a documented, low-real-cost concession package satisfies their scoreboard without margin loss. Operational rule: **never reduce price without removing scope.**
The table below is a lookup, not a menu - each row is keyed to the move the buyer makes, so ranking the counters would be false precision: you never choose between them.
| Buyer move | Mechanism | Counter |
| ------------------------------------ | --------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------- |
| "You'll have to do better than that" | Vague pressure inviting you to cut against yourself | Ask what specifically must improve, and by how much; trade, do not move |
| Budget bogey | Friendly fixed-budget anchor that makes you solve their problem | Confirm who decides and who pays; fit scope to the number; present a pre-built MESO at that budget |
| Claimed limited authority | "I can't approve that" buys time and free movement | Mirror it - your pricing is set above your level too; withhold firm concessions until their authority is present |
| Competitor price claim | Real or bluffed cheaper quote | Ask for the quote in writing; price the scope difference instead of matching the number |
| Unbundling / line-item pricing | Exposes where margin concentrates | Re-price components so the bundle discount disappears when unbundled |
| Timing to your quarter-end | Buys your calendar-driven desperation | Anchor timeline on their compelling event; cap end-of-quarter discount authority |
## Measurement
Plan readiness - the plan is not done until all of these pass; iterate until 100%:
- Every listed concession has (a) a named reciprocal get, (b) a cost-to-us magnitude, not a currency figure, (c) the approval level required - including the review a compliance-exposed lever triggers.
- Rows sit in ratio order within each tier, and the levers the effort ceiling ruled out are struck and named, not sitting at the bottom of a tier.
- The walk-away line is stated in writing and derived from the alternative, not the margin floor.
- Every planned trade has its conditional sentence written out.
- At least two MESO packages exist, roughly equal in value to the seller.
Outcome KPIs to track after the negotiation:
- Concession-to-get ratio: share of granted concessions with a documented reciprocal commitment on the order form.
- Realized price vs list, per deal and per segment.
- Discount distribution shape: bunching just under an approval threshold means the matrix is being gamed, not respected.
- Win rate and cycle time by discount band - confirms price gains are not buying losses elsewhere.
- B2C: front-end vs back-end gross per unit, and retained vs booked (a deal can post strong gross and lose it to refunds and early payoffs).
## References
- mbfinotti/sales-skills@deal-value-calc - quantify the deal's underlying value and ROI (the value-to-them inputs for step 3)
- mbfinotti/sales-skills@sales-objection-handling - in-call rebuttals to price and competitor objections
- mbfinotti/sales-skills@deal-champion-mapping - map which stakeholder receives which concession
- mbfinotti/sales-skills@sales-meeting-recap - paper the granted trades and their gets after the call