# Offer Card template Part of [Offer Lab](/free/offer-lab/) · By Aymane Hitmi · Labgenz · Last reviewed: 9 November 2026 The Offer Card is one page that holds your chosen offer, the evidence behind it, the price you'll test and the rules that decide whether it survives. It isn't a pitch. It's a hypothesis written so the market can reject it. Copy the card below into a file called `offer-card.md` in your workspace folder. Fill it from Prompts 5, 6 and 7 of the chain. Tag every field **[DECISION]** (a choice you make: offer name, promise, boundary, price, structure), **[FACT: IDs]** (backed by the evidence, edge or reach IDs in the tag), **[ASSUMPTION]** (you believe it, but haven't checked) or **[GAP]** (you don't know yet). A decision is never a fact, however sure you are. The assumptions and gaps are your interview plan. ## The card ```text OFFER CARD v1 — {{YOUR_NAME}} — {{DATE}} Status: DRAFT / TESTING / CONTINUE / RESHAPED / KILLED 1. OFFER Offer name (plain words): One line: For [specific buyer] who [problem in their words], I deliver [outcome I control] in [timeframe], using [mechanism]. 2. BUYER Who exactly (3+ observable traits): Where I find 10 of them this week (R-IDs): Who pays (if different from who has the problem): Buying trigger (the moment they'd look for this): 3. PROBLEM Their problem, in their words (1-2 quotes + E-IDs): What it costs them today (consequence, E-IDs or ASSUMPTION): What they do or pay today instead (E-IDs or GAP): Alternatives I checked (tool, feature or service + where I saw it): 4. PROMISE Outcome I control: Timeframe: Boundary (what I do NOT do or promise): 5. DELIVERY What they receive (3-5 things they could point at): - - - How I deliver it (tag each step AI / ME / BUYER, mark CHECK steps): - - - - What I need from the buyer (access, material, time): My hours per client: Tool costs per client: 6. WHY ME Edge + proof (K-IDs): 7. PRICE HYPOTHESIS All prices excluding VAT / sales tax: yes Floor (hours incl. onboarding, handover, share of sales calls x minimum rate + tools + payment fees, then + 20% buffer): Ceiling (buyer's current alternative, or "unquantified"): Market anchors (only prices I've actually seen, with source): Test prices: LOW ____ / ANCHOR ____ / HIGH ____ Currency: Basis for each (floor, ceiling, anchor, E-ID, or PLACEHOLDER + the call question that will replace it): Structure (one-off or monthly; deposit; what costs extra): Pilot version (smaller scope, not a discount): Pilot floor (same formula, pilot hours) and pilot price: Reasoning (4 sentences, no ROI claims): 8. SCORES (from my sheet, after my changes) Pain __ / Frequency __ / Pay __ / Reach __ (x2 = __) / Edge __ Total __ /30 · Evidence-backed __ /3 (Pain, Frequency, Pay citing an E-ID) Gate IN / THIN / OUT Runner-up and why it lost: 9. TEST PLAN Top 3 assumptions, riskiest first: A1: A2: A3: Conversations: 10 by {{DEADLINE}} via {{CHANNELS}} Kill criteria (countable, each mapped to ONE of KILL / RESHAPE / CONTINUE, listed in check order: "fewer than 10 conversations" first, then KILLs, then RESHAPEs, then CONTINUE): - - - - - ``` ## How to fill each section ### 1. Offer **Offer name.** What a buyer would call it when telling a colleague. Weak: "AI Growth Accelerator". Strong (fictional): "Reschedule and Waitlist Kit". **One line.** If you can't fill all four brackets, you don't have an offer yet. You have a topic. The outcome must be something you deliver and control, never the buyer's revenue or growth. ### 2. Buyer **Who exactly.** Three or more traits you could check from the outside: trade, size, situation, tools they use, who does the work today. Weak: "small businesses that want to grow". Strong (fictional): "independent physio clinics, 2 to 6 practitioners, owner still runs the schedule, bookings by phone and WhatsApp". **Where I find 10 this week.** Name the channel and the reach ID. If you can't, your Reach score was too high. **Who pays.** The person with the problem isn't always the person with the budget. A receptionist feels the pain; the owner signs. Sell to the payer, quote the person who feels it. **Buying trigger.** The event that makes them search: a staff member leaving, a bad review, a new season, a compliance date. Offers with a clear trigger are easier to time your outreach around. ### 3. Problem Use the buyer's words, copied, with the evidence ID. Your paraphrase is a belief; their sentence is evidence. If "what they pay today" is a gap, it's the first thing to ask in every call. List the alternatives you checked at your desk (the features of tools they already use, services they could hire) with where you saw each; never list one you haven't seen yourself. ### 4. Promise Outcome, timeframe, boundary. All three, every time. - **Outcome I control:** a working system, a set of templates, a document, a setup. Not "more clients", "more bookings" or "hours saved". Those depend on their market and their team, and promising them is a trust risk and a legal one. - **Timeframe:** in working days or weeks, counted from the moment you have what you need from them. - **Boundary:** what you won't do. "I set it up and hand it over. I don't answer your customers for you." Boundaries prevent scope creep and make the offer easier to say yes to. ### 5. Delivery **Deliverables** are nouns. "12 reply templates", "a waitlist sheet", "a one-page procedure", "a 45-minute handover call". If a deliverable is a feeling ("peace of mind") or a category ("strategy", "support"), rewrite it. **Steps** show where AI does the work and where you check it. A delivery plan with no CHECK step is a plan to ship unreviewed AI output to a paying client. Mark at least one CHECK before anything reaches the buyer. **What I need from the buyer** is where most first deliveries stall. Be specific: exports, logins (set up by them, in their account, never shared in a chat), past examples, a decision-maker's time. If they won't give you this in the sales conversation, they won't give it to you after paying either. ### 6. Why me One or two proofs from your edge table: the job you did, for how long, in which industry, and something you could show. "I ran the front desk at two clinics for eight years" (fictional) beats "I'm passionate about healthcare". ### 7. Price hypothesis Three reference points, then three test prices. - **Floor:** what delivery costs you, with a buffer. Count every hour the client costs you: onboarding, delivery, handover and your share of the sales calls it took to win them. Add payment fees, and state every price excluding VAT or sales tax (in the EU, a VAT-registered seller adds VAT on top; say which applies to you). You never test below the floor. - **Ceiling:** what the buyer's current alternative costs them, from evidence only. If you can't quantify it, write "unquantified" and ask in your calls. - **Anchors:** prices you have actually seen for similar offers, with the source. Never use a "typical rate" an AI gave you from memory. - **Basis:** every test price gets one line saying where it comes from. If you have no ceiling and no anchors yet, label the number PLACEHOLDER and name the call question whose answers will replace it. - **Test prices:** say the ANCHOR in conversations, note the reaction word for word, and move the number based on what buyers do (accept a pilot, pay a deposit), not on what they say. - **Pilot:** a smaller scope at a smaller price, not a discount on the full scope. It gets its own floor from its own hours, and its price sits above that floor. Discounts teach buyers your price is negotiable; a pilot teaches them what working with you is like. ### 8. Scores Copy them from your sheet after your own changes, not from the model's first draft. Evidence-backed counts only Pain, Frequency and Pay scores that cite buyer evidence (an E-ID), from 0 to 3. Reach and Edge come from your own notes about yourself, so they never count. Fewer than 2 is THIN, whatever the total, and the card should say so. ### 9. Test plan **Assumptions** are ranked by how much damage they'd do if wrong. The riskiest one is usually about payment ("they'll pay an outsider for this") or access ("they'll share their data"). **Kill criteria** are numbers you can count after ten conversations, each tied to exactly one decision. Write them before your first call, and don't move them afterwards. If you change them, write a new card version instead. **Check them in order**, so you can't pick the result you like. First, "fewer than 10 conversations": if it fires, re-score Reach and finish the conversations before deciding anything. Then any KILL that fires beats any RESHAPE, and any RESHAPE beats CONTINUE. CONTINUE only counts when nothing else fired. If someone already booked a pilot, deliver what you agreed, then follow the decision. ## Validation log Add this as a tab in your sheet, one row per conversation. Fill it within an hour of each call, while the words are fresh. | # | Date | Channel | Buyer type | Problem confirmed unprompted (Y/N) | Instance in last 30 days (Y/N) | Current workaround or spend | Reaction to anchor price (verbatim) | Highest commitment | Quote worth keeping | |---|---|---|---|---|---|---|---|---|---| | 1 | | | | | | | | | | | 2 | | | | | | | | | | | 3 | | | | | | | | | | **Highest commitment** is one of: none, material (they shared real, anonymised material), intro (they introduced a peer), pilot (they booked a pilot), deposit (they paid). Compliments are not on the list. ## Decision log After the deadline, write one entry. Keep old cards; never overwrite them. ```text DECISION — {{DATE}} — Offer Card v{{N}} Conversations held: __ /10 Kill criteria triggered, in check order: (list, or "none") Decision: CONTINUE / RESHAPE / KILL What changes in v{{N+1}} (if reshaping): What I learned that I didn't know when I wrote v{{N}}: ``` ## Before you call it done - [ ] Every field is filled and tagged [DECISION], [FACT: IDs], [ASSUMPTION] or [GAP]. - [ ] The name, promise, boundary, price and structure are tagged [DECISION], not [FACT]. - [ ] The one line has all four brackets filled. - [ ] The promise has no income, revenue, bookings or time-saved claim. - [ ] At least one delivery step is a CHECK step. - [ ] The price is stated excluding VAT, has a floor that includes onboarding, handover, sales time and payment fees, a basis line for every test price, and no invented market rates. Any pilot clears its own floor. - [ ] The kill criteria are countable, cover all three assumptions, map to one decision each, and are listed in check order. - [ ] No names, emails, phone numbers or client data appear anywhere on the card.