The Buyer Persona and ICP Reboot — 60-Min Training

Direct Answer

This is a runnable script. Print it, hand it to your sales manager, and follow the clock. Total run time: 60 minutes flat. Room setup: whiteboard, sticky notes, and the last 20 closed-won and 20 closed-lost CRM records on a shared screen.

Section 1 — Opening Frame and the "Persona Is Not a Job Title" Reset (5 min)

Open cold. Read this verbatim to the room:

*"Raise your hand if your current persona doc says 'VP of Sales, 50-500 employees, SaaS.' Keep it up if that has ever, even once, made you book a meeting. Hands down. That is not a persona. That is a LinkedIn filter. Today we fix that."*

• Define the gap out loud: A persona is a *Job-to-be-Done plus a trigger event*, not a title plus a company size. Tony Ulwick's framing in *Jobs to be Done: Theory to Practice* (2016) is the spine — buyers "hire" software to make progress on a job, and the job exists whether or not your product does.
• Set the deliverable: By 4 PM the team produces (1) a one-page ICP score card, (2) two persona one-pagers, (3) an anti-ICP list of 25 accounts to stop touching.
• Name the cost: Reference SiriusDecisions / Forrester's finding that 70% of deals lost are lost to "no decision," not to a competitor — the root cause is almost always a persona-trigger mismatch.

Section 2 — Build the ICP Score Card: Firmographic + Technographic + Trigger (15 min)

Pull up the last 20 closed-won accounts. On the whiteboard, draw three columns.

1. Firmographic column. For each won deal, write employee count, ARR band, geography, funding stage, and industry sub-vertical. Look for the cluster, not the outlier. Aaron Ross's *Predictable Revenue* (2011) ICP method is column one: what does the company look like on paper?
2. Technographic column. What is in their stack? Salesforce vs HubSpot, Snowflake vs Redshift, Outreach vs Salesloft. Use BuiltWith or Wappalyzer evidence pulled before the meeting. This is the column that separates an okay ICP from a great one — Bowery Capital's ICP work with portfolio companies repeatedly shows tech stack is a stronger predictor of fit than industry.
3. Trigger-event column. What changed in the 90 days before they bought? New VP hired, Series B raised, acquired a competitor, missed a quarter, ISO-27001 audit failed. Triggers are *why now,* and "why now" is what closes deals.

Score each criterion 0-3. Sum to a 0-30 ICP score. Anything ≥ 22 is in. Anything ≤ 12 is anti-ICP and goes on the do-not-prospect list. Write the rubric on the wall and photograph it before erasing.

• Worked example to read aloud: *"Stripe-stack, 150-employee Series B vertical SaaS company in fintech, just hired a Head of RevOps in the last 60 days = ICP 28. Bootstrapped 8-person agency on QuickBooks with no RevOps function = ICP 7. We do not prospect 7s."*

Section 3 — Persona JTBD Overlay: The Functional, Emotional, Social Stack (10 min)

Now the human inside the company. Hand each AE a blank persona one-pager. They fill three boxes — straight from Ulwick's Outcome-Driven Innovation:

• Functional Job: what the buyer is literally trying to accomplish. *"Forecast next quarter within 5% of actuals."* Not "use a CRM."
• Emotional Job: what they want to *feel.* *"Stop being the person the CEO publicly blames in the QBR."*
• Social Job: how they want to be *seen.* *"Be the operator the board notices. Get promoted to CRO inside 18 months."*

Then a 90-second pains/gains pass borrowed from Adele Revella's *Buyer Personas* (2015) "5 Rings of Buying Insight": Priority Initiatives, Success Factors, Perceived Barriers, Decision Criteria, Buyer's Journey. Write one sentence per ring. If an AE cannot fill all five from memory for their top persona, they have a homework assignment due Friday — three discovery calls and a rewrite.

Section 4 — The Win-Loss Interview as Persona Source of Truth (10 min)

Stop guessing. Go to the data. Read this script verbatim to your last five closed-won buyers and last five closed-lost buyers. 15 minutes per call, recorded, transcribed.

• *"Walk me back to the week you first realized this was a real problem. What happened that week?"* — surfaces the trigger.
• *"Who else was in the room when you made the decision? What was their biggest objection?"* — surfaces the buying committee and the killer.
• *"What were you using before, and what specifically broke?"* — surfaces the displaced solution and the switching cost.
• *"If you had not bought us, what would you have done in the next six months?"* — surfaces the real competitor (which is almost always "do nothing").
• *"What did we almost lose the deal on?"* — surfaces the buying criterion you under-served.

Andy Raskin's *The Greatest Sales Deck I've Ever Seen* (2016) is built on this exact discipline — the "Name the Enemy" opening of a sales narrative comes directly from win-loss transcripts, not from product marketing. If your narrative does not echo language you heard in five buyer interviews, it is fiction.

Section 5 — Build the Anti-ICP and Map Trigger Events to Cadence (15 min)

This is the section most teams skip. Do it.

• Anti-ICP list: Each AE names five accounts currently in their pipeline that score ≤ 12. Manager pulls them out of the forecast on the spot. The room will get quiet. That is the point. Dave Gerhardt and the Bowery Capital playbook both emphasize that subtraction is the highest-leverage move in early-stage GTM.
• Trigger-to-cadence map: For each high-value trigger event, write the first-touch message. Examples:
• *Trigger: New VP RevOps hired in last 45 days.* Day-1 SDR email: *"Saw you just took the RevOps seat at [Co]. The first 90 days usually break on forecast accuracy or attribution — which one is on your desk this week?"*
• *Trigger: Series B announced.* Day-1 SDR LinkedIn: *"Congrats on the B. The companies that hit their next round usually rebuild the GTM stack in the first 6 months — happy to share what the last four [vertical] Series Bs did, no pitch."*
• Pirate Metrics (AARRR) tie-in: Dave McClure's framework — Acquisition, Activation, Retention, Referral, Revenue — is the dashboard the new ICP score card *feeds.* If acquisition volume drops but activation rate climbs, the ICP rewrite is working. Watch activation-rate-per-ICP-score weekly for the next 90 days.

Section 6 — Commitments, Homework, and Close (5 min)

Lock it in. Around the room, each rep states out loud:

• One account they are firing from their pipeline this week.
• One trigger event they will set up an alert for in LinkedIn Sales Nav or Common Room before Friday.
• One win-interview they will book in the next 10 business days.

Manager closes:

*"The score card on the wall is the new pipeline gate. Nothing under 22 goes in the forecast starting Monday. We will rebuild this score card every 90 days off the next 20 closed-won. The persona doc is a living document, not a PDF in a Drive folder."*

Photograph the whiteboard. Drop it in the team Slack with the title "ICP Score Card v1 — expires 2026-08-26." Calendar the 90-day rebuild now, in the meeting, before anyone leaves.

FAQ

Q: How is "persona" different from "ICP"? A: ICP describes the *company* you sell to (firmographic + technographic + trigger). Persona describes the *human* inside that company (functional + emotional + social JTBD). You need both. An account can be perfect ICP and the persona is wrong, or vice versa — either way the deal dies.

Q: How many personas should we have? A: Two to four. More than four and the team cannot remember them, which means they are not real. Adele Revella's research across 100+ B2B companies in *Buyer Personas* (2015) finds three is the modal number that ships.

Q: What if win-loss interviews are awkward to request? A: Open with: *"I'm not selling you anything — I am rebuilding our persona research and you are one of five buyers whose insight matters most. 15 minutes, I'll send the transcript."* Conversion rate on that ask is typically 60-70% inside 90 days of close.

Q: How often should we rebuild the ICP score card? A: Every 90 days against the last 20 closed-won. Faster than that and you chase noise. Slower and the market drifts past you — especially in B2B SaaS where buying committees and tech stacks shift quarterly.

Q: What if leadership refuses to cut the anti-ICP accounts from forecast? A: Run the math. Pull the average days-in-stage and close rate for accounts scoring ≤ 12 vs ≥ 22 over the last four quarters. The gap is usually 4-6x on close rate and 2-3x on cycle length. Leadership cuts the list when the spreadsheet does the arguing for you.

Sources

1. Ulwick, Anthony W. *Jobs to be Done: Theory to Practice.* Idea Bite Press, 2016. — the canonical JTBD methodology and Outcome-Driven Innovation framework.
2. Revella, Adele. *Buyer Personas: How to Gain Insight into Your Customer's Expectations, Align Your Marketing Strategies, and Win More Business.* Wiley, 2015. — the "5 Rings of Buying Insight" persona research method.
3. Ross, Aaron and Marylou Tyler. *Predictable Revenue.* PebbleStorm, 2011. — the original outbound ICP scoring approach.
4. Raskin, Andy. "The Greatest Sales Deck I've Ever Seen." Medium, 2016. — Name-the-Enemy narrative construction from win-loss data.
5. McClure, Dave. "Startup Metrics for Pirates (AARRR)." Master of 500 Hats, 2007. — Pirate Metrics framework for measuring ICP-rewrite impact.
6. SiriusDecisions / Forrester. *B2B Buying Study,* multi-year — the "70% lost to no-decision" benchmark.
7. Bowery Capital. *Early-Stage Customer Development & ICP Playbook,* portfolio operating notes — technographic-as-predictor finding.
8. Gerhardt, Dave. *Founder Brand.* 2022 — subtraction-as-leverage GTM principle for early-stage teams.