Accounts Don't Buy. People Do.  🔗 **Source:** [Accounts Don't Buy. People Do.](https://www.youtube.com/watch?v=cKDamhLwow8) > [!note]+ Document Notes > tags: [abm, buying-committee, b2b-sales, go-to-market, sales-marketing-alignment, devrel-measurement] > author: [Common Room webinar — Will Taylor with Mason Cosby (Scrappy ABM) and Davis Potter (ForgeX)] > source: [YouTube — Common Room] > reference: [https://www.youtube.com/watch?v=cKDamhLwow8] > date created: 2026-08-07 00:00 > > > [!info] Generated analysis — not the author's words. Sections below are machine-written > > from the captured source. Only text inside `> [!quote]` blocks is verbatim from the author. > > Do not quote or attribute any other line to the named author. > > # 🏷️ Title: Accounts Don't Buy. People Do. (Common Room webinar, full session) > <!-- deep-read:sections start --> ## 🧭 Table of Contents - [🔗 References](#-references) - [📝 Summary](#-summary) - [📖 Main Content](#-main-content) - [📚 References](#-references-primary) - [🗣️ Paraphrase](#️-paraphrase) - [📰 Newsletter](#-newsletter) - [✅ Action Items](#-action-items) - [🧪 Hypotheses](#-hypotheses) - [📊 Methodology](#-methodology) - [❓ Questions](#-questions) - [🔮 Future Research](#-future-research) - [💡 Implications](#-implications) - [🧠 Learning Styles Assessment](#-learning-styles-assessment) - [📊 Accessibility Assessment](#-accessibility-assessment) - [🌍 DEIB Assessment](#-deib-assessment) - [✅ Logical Validity](#-logical-validity) - [🛡️ DARVO Analysis](#️-darvo-analysis) - [🎯 Conclusion](#-conclusion) - [🧱 Refactor Appendix](#-refactor-appendix) ### 🔗 References - Video: https://www.youtube.com/watch?v=cKDamhLwow8 (published 2026-08-06, ~5,869 transcript words) - **Companion document in this library:** the Common Room blog recap of this same session — `01kze5620hqprv2zdxwcycsev0`, https://www.commonroom.io/blog/abm-webinar-recap/ - Scrappy ABM (Mason Cosby) · ForgeX (Davis Potter) · Common Room (host and publisher) - SiriusDecisions Demand Waterfall — referenced in-transcript as "serious decisions as demand waterfall" (ASR) > [!important] **This is the primary source; the blog post is its recap.** Read them as a pair. > The webinar contains **material evidence the recap omits** — most importantly, a confound that > undercuts the recap's headline statistic. See Logical Validity → fallacy 1. Where the two > disagree, this transcript is the original. ## 📝 Summary - **What happened.** The full ~45-minute webinar behind the Common Room recap. Cosby and Potter argue that account-level ABM scoring cannot tell sales who to contact, and that the unit of action must be the individual buyer inside an eight-role buying committee. - **Why it matters.** Beyond the framework, the session supplies three things the recap dropped: the **confound** explaining why fast deals are fast, a concrete story about buying insurance to remove a blocker, and a striking base rate — roughly **5% of the 338 organizations Cosby spoke with were running any marketing into pipeline at all**. - **Key stakeholders.** B2B marketing/sales leaders; RevOps; DevRel and community teams under pressure to attribute pipeline; anyone whose signal data resolves to accounts rather than people. - **Risks and tradeoffs.** The evidence is one practitioner's self-reported book, the arithmetic does not reconcile (see below), and the headline cycle-time comparison compares two structurally different kinds of deal. The framework is sound; the proof is not. - **Bottom line.** Better than its recap. The confound Cosby volunteers in passing is the most intellectually honest moment in either document, and it happens to demolish the causal reading the recap promotes. ## 📖 Main Content - **The trust-collapse mechanic (verbatim origin).** The line the recap quotes originates here: an account is hot, sales asks why, > [!quote] marketing gets to respond back with, "Well, it's 80 points." And at that point, um sales immediately loses trust, doesn't actually know who to reach out to, and everything kind of falls apart in that exact moment. Cosby attributes the failure specifically to "the broken promise of ABM" from "our big three tech vendors from 2020." - **Potter's historical arc.** MQL hamster wheel (SiriusDecisions demand waterfall era) → ABM unifies teams behind a target account list → new failure mode, the MQA thrown over the wall → current push toward buying groups. His framing throughout: identify, **validate**, and continuously **re-engage** the group in partnership with sales. - **Explicitly not either/or.** Potter is emphatic: > [!quote] what I don't want people to walk away with today is thinking let's completely throw out the account-based approach and just go to buying groups. It's not a it's not a eitheror. The recommendation is a blend, with the blend ratio left undefined. - **The dataset, with its own arithmetic problem.** Cosby states 338 sales cycles over three years, 118 closed won, and the remainder as "20 whatever the remainder 211" closed lost. **118 + 211 = 329, not 338.** He later refers to "339 whatever sales cycles." The numbers are spoken approximations, not a reconciled dataset — worth knowing before anyone cites them. - **The eight roles, with sharper definitions than the recap gives.** Decision maker (says yes); economic buyer (finance — not the go/no-go, but the terms negotiation); champion (drives internally); key influencers (manager level); technical/implementation ("we get most often blocked by IT"); blockers (> [!quote] they aren't vocal on the call. They just sew seeds of doubt behind the scenes that you never get to actually directly address unless your champion's amazing and brings them to you); end users (matter at retention, not first sale — "if nobody uses the product then you're going to not get the second, third, fourth, seventh sale"); gatekeepers, which he distinguishes from blockers precisely — **gatekeepers protect time but hold no opinion**, executive assistants being the archetype. - **The insurance story — the best concrete artifact in the session.** Cosby's agency kept losing deals at the IT-review stage. Someone eventually told them they could not sign because the agency had no cyber insurance. He bought a policy (he corrects himself mid-sentence: $4,000, then "actually, sorry, it's $14,000 a year") and the blocking stopped. His framing: > [!quote] you lose so many deals, you do so much work to then lose over a freaking $14,000 a year insurance policy. Just buy it regardless. This is the clearest illustration in either document that blockers are often *procedural*, not persuasive — and that the fix is operational rather than rhetorical. - **The 5% base rate.** Of the ~338 organizations he ran deep-dive discussions with, > [!quote] Roughly 5% of them were running any kind of marketing in pipeline. His inference: sales is limited to whoever the champion grants access to, whereas marketing can reach the economic buyer and decision maker directly with ads. This is the actual argument for marketing's role in the committee, and the recap omits it. - **The cycle-time claim — and its confound.** Deals with more committee engagement earlier average 60 days; those engaged through marketing before booking the meeting run 30 days. Then Cosby adds the line the recap leaves out: > [!quote] They're also not exploring other vendors or competitors. He goes on to contrast it explicitly with "here's the RFP and you're going to be one of five vendors" — calling them "totally different" sales. He is describing two different deal types, not one deal type accelerated. - **Forecasting benefit.** He claims the mapped committees let him forecast the next quarter more confidently, and stresses the timeline honestly: "it takes 12 months. Like it's not going to happen overnight." - **Enterprise complexity (Potter).** Buying groups vary by industry, product line, org maturity, and business unit. Title aggregation misleads inside large banks whose siloed units hold competing vendor contracts. A concrete audience example: "civil engineer" denotes entirely different roles across organizations, so title-based enrichment breaks. - **Discovery stack when sales knows little.** (1) Instrument cheaply — call recorder ~$20/month plus HubSpot Starter ~$20/month, sales extension BCCs every email to the contact/account record, revealing who sales engages *and who they do not*. Caveat he states plainly: you then wait a year, because "you can't record a call that wasn't recorded." (2) Internal interviews with sales, CS, and enthusiastic customers about champion heroics and blockers, then ask whether each story is "an anomaly or a standard." (3) AI via the **CRIT** framework — Context, Role, Interview, Task — where the model interviews *you* to fill gaps before producing a committee map you then validate against real data. - **Format note.** Will Taylor hosts explicitly for audience participation, repeatedly inviting chat questions and offering to bring attendees on stage; at least one audience question (from "Carlo") is answered live. Will also volunteers that his own buying-group figure was 6–10 against Potter's 12–14, and treats Potter's number as validating an upward trend rather than contradicting him. ### 🗒️ Notes - Will opens by citing his own writing that **trust between buyers within a buying group is eroding**, not just buyer trust in vendors — an interesting claim raised and never developed. - Cosby positions his work as "unintentional research": he sees what makes clients money and standardizes it. Honest about being practice-derived rather than designed study. - ASR corruptions: "Forjax" = ForgeX, "Scrap ABM" = Scrappy ABM, "Conorman room" = Common Room, "serious decisions" = SiriusDecisions, "sew seeds of doubt" = sow seeds, "crit"/"CR" = CRIT, "EAS" = EAs (executive assistants), "AVM" = ABM. - The banter is heavy — repeated hand-offs between speakers, laughter, a running bit about deviating from the doc. Roughly the first three minutes carry no content. ## 📚 References (primary) - The companion recap: https://www.commonroom.io/blog/abm-webinar-recap/ (`01kze5620hqprv2zdxwcycsev0` in this library) - LinkedIn B2B buying-group research (12–14 figure, cited without link in both documents) - SiriusDecisions Demand Waterfall — the MQL-era model Potter positions as the prior paradigm - MEDDIC / MEDDPICC and The Challenger Sale — established prior art on economic buyer, champion, and mobilizer/blocker taxonomies, unacknowledged in the session ## 🗣️ Paraphrase - **The setup.** Most ABM looks only at account-level data, which is not how buyers actually buy, and buying groups keep growing. - **The failure.** A score cannot answer "who do I call," so handing sales a hot account destroys trust rather than creating alignment. - **The history.** MQLs threw individuals at sales; ABM aligned teams around accounts and then threw accounts over the wall instead. - **The correction.** Blend account-level prioritization with person-level action; do not swing the pendulum. - **The map.** Eight recurring roles, with gatekeepers distinguished from blockers by the absence of an opinion. - **The unlock.** Marketing can reach committee members sales cannot, because sales only meets whoever the champion permits. - **The evidence.** Committee-engaged deals move faster — though the fast ones are also the ones with no competing vendors. - **The method.** Instrument cheaply, interview internally, use AI to hypothesize, validate with sales, expect it to take a year. ## 📰 Newsletter **The $14,000 insurance policy that was costing them millions** Common Room's "Accounts Don't Buy. People Do." webinar has a story in it that every DevRel and marketing person should hear. Mason Cosby's agency kept losing deals. Not on price, not on fit — they'd get deep into the cycle and stall at IT review. Eventually someone told him the actual reason: you don't carry cyber insurance, so we can't sign. He bought a policy. Fourteen thousand a year. The blocking stopped. That's the whole argument for mapping a buying committee, told better than any framework diagram. The person killing your deals often isn't arguing with you — they're applying a checklist you never saw. Cosby's line: "you lose so many deals, you do so much work to then lose over a freaking $14,000 a year insurance policy. Just buy it regardless." The other number worth your attention: of the ~338 companies he's run deep-dive sales conversations with, roughly **5%** were doing any marketing into an open pipeline. Sales only ever meets who the champion lets them meet. Marketing can run ads at the economic buyer directly. That's the actual case for marketing's role in a deal, and it's more persuasive than the funnel diagrams. One honesty note, because it matters. The recap of this session says early committee engagement halves the sales cycle, 30 days versus 60. In the video Cosby adds the part the recap dropped: the 30-day deals "are also not exploring other vendors or competitors." Those aren't accelerated deals. They're sole-source deals versus five-vendor RFPs — his words, "totally different." Great framework. Don't cite the stat. **Links:** [the webinar](https://www.youtube.com/watch?v=cKDamhLwow8) · [the recap](https://www.commonroom.io/blog/abm-webinar-recap/) ## ✅ Action Items - **DevRel** - `#action/h1` Find your insurance policy. Ask lost/stalled adopters what *procedural* thing blocked them — a missing SOC 2, no SBOM, an unclear license, no security contact. Procedural blockers are cheap to fix and invisible until asked about directly. - `#action/h1` The 5% statistic reframes DevRel's value proposition: sales reaches only who the champion permits; content and community reach the roles sales never meets. That is a defensible, non-vanity argument for DevRel investment. - `#action/h2` Replace score-shaped reporting with named-state reporting, per the trust-collapse mechanic. - `#action/h2` Distinguish gatekeepers from blockers in adoption work — one needs befriending, the other needs their objection surfaced and answered. - `#action/h3` Run the closed-lost interview round; ask specifically what the champion had to do internally. - **OSPO** - `#action/h1` Map your own approval committee for inbound OSS adoption: legal, security, architecture, procurement. The insurance story is precisely an OSPO-shaped blocker being resolved with a purchase rather than an argument. - `#action/h2` Publish the artifacts that unblock technical review before you are asked — security policy, SBOM, license clarity. That is the OSS equivalent of buying the policy. - `#action/h3` Note the siloed-business-unit warning; large orgs frequently adopt the same OSS twice with no shared record. - **Community** - `#action/h2` End users determine renewal, not first adoption — the argument for investing in post-adoption community rather than acquisition. - `#action/h3` "Who are we not talking to" as a standing coverage question for community programs. - **WordPress (by team)** - Marketing: enterprise WordPress adoption is a committee decision; account-level interest data cannot tell you whether the blocker is security review or the implementation team. - Hosting: the most directly commercial team and the one where this maps unchanged. - Support: end-user experience determines renewal and expansion, which is the retention case for support investment. - Training: enterprise training purchases are committee decisions with distinct economic buyers and end users. - Community: champion/blocker/gatekeeper taxonomy applies to sponsor and contributor recruitment. - Meta / Documentation: publishing security and compliance artifacts removes procedural blockers for enterprise adopters — the direct analogue of the insurance purchase. - Core, Core Performance, Core AI, Core Program Team, Mobile, Playground, Design, Photos, TV, Polyglots, Accessibility, CLI, Tide, Openverse, Themes, Plugins: Not applicable to this source — a B2B sales-motion webinar offers these teams nothing actionable. - **The WPOCC** - `#action/h1` Sponsorship pursuit is a committee sale. Map economic buyer (finance), champion (the sympathetic staffer), blocker (legal/brand), and gatekeeper (the EA) inside target sponsor orgs. The insurance lesson applies directly: identify the procedural blocker preventing a sponsorship and remove it rather than making a better pitch. - **FAIR** - `#action/h2` Institutional adoption of federated infrastructure will be a committee decision, not a developer decision. Publish the artifacts security and procurement will demand before they ask. ## 🧪 Hypotheses 1. The 30-day cohort differs from the 60-day cohort by competitive context (sole-source vs RFP), not by marketing engagement. *Cosby's own aside supplies the evidence for this; testable by segmenting his cycles on competitive presence.* 2. Procedural blockers (missing certifications, insurance, compliance artifacts) account for a larger share of losses than persuasion failures. *Testable from CRM loss reasons, and the insurance anecdote is a strong prior.* 3. Fewer than 10% of B2B organizations run marketing into open pipeline. *Cosby's ~5% figure, generalizable and testable via survey.* 4. Title-based enrichment has a high role-misassignment rate in large multi-business-unit organizations. *Potter's "civil engineer" example; testable against verified org charts.* 5. Gatekeeper-directed effort (befriending EAs) has measurable effect on meeting-set rates independent of champion strength. *Testable via A/B in outbound.* 6. Committee-mapping improves forecast accuracy, as Cosby claims for his own Q2. *Testable by comparing forecast error before and after mapping.* ## 📊 Methodology **Of the source:** weak, and weaker than it sounds. This is a vendor-hosted webinar with two speakers who sell adjacent services. The evidence base is Cosby's self-reported agency book: 338 cycles over three years, no control group, no segmentation disclosed, definitions ("more folks engaged," "earlier") left operationally vague. **The stated figures do not reconcile** — 118 won plus 211 lost is 329, against a stated total of 338, with a later reference to 339. Potter contributes framing and cautions but no data. The eight-role taxonomy is derived from won deals. The one genuinely useful base rate (5% running marketing in pipeline) comes from recalled sales conversations, not a survey instrument. **Of this analysis:** full transcript read from the Reader `notes` capture (~5,869 words, auto-generated captions). All claims trace to explicit transcript statements. The arithmetic inconsistency was computed directly from the spoken figures. The comparison against the companion recap (`01kze5620hqprv2zdxwcycsev0`) is a **cross-document check I performed** — the confound is present in this transcript and absent from the recap; that is a verifiable difference between two documents in this library, not an inference about intent. ## ❓ Questions - What is the competitive-context split between the 30-day and 60-day cohorts? This single number determines whether the central claim survives. - Which of 338/329/339 is the real denominator, and does the win-rate figure change materially? - What share of the 211 losses were also multi-threaded? Without it, threading tells us nothing. - How much of the 5% figure is selection — do companies already running marketing-in-pipeline simply not hire an ABM agency? - What is the CRIT map's error rate before sales validation, and what does acting on a wrong map cost? - Where is privacy? Neither document mentions GDPR/CCPA despite the entire method being person-level identification from behavioral signal. - Will's claim that trust *between buyers inside a group* is eroding was raised and dropped — what is that based on? ## 🔮 Future Research - Segment the cycle-time comparison by competitive context; this is the study that would settle the article's headline claim, and the data already exists. - Quantify procedural versus persuasive loss causes across a multi-vendor sample. - Test whether the eight-role taxonomy generalizes to open source adoption and to grant/sponsorship decisions. - Measure AI-generated committee-map accuracy against sales-verified ground truth. - Investigate intra-committee trust erosion — the most novel unexplored claim in the session. ## 💡 Implications - **Practice.** Ask lost deals what procedurally blocked them. The answer is often cheap, specific, and fixable — and nobody volunteers it unprompted. - **Policy.** Person-level identification from behavioral signal is a governance question. Its total absence from both documents is itself a finding about the category. - **Security.** Inverted relevance: security review *is* the blocker in this story. Publishing compliance artifacts early is a go-to-market act, not just a security one. - **Education.** The gatekeeper/blocker distinction — opinion versus no opinion — is a genuinely useful teaching primitive. - **Funding.** Applies directly to sponsorship and grants: committees decide, procedural blockers kill, and the champion who cannot introduce you is not a champion. ## 🧠 Learning Styles Assessment - **Visual.** Poorly served in the transcript; the session presumably had slides, none of which survive. The eight roles are delivered as a spoken list, which is hard to hold. - **Aural.** Strongly served — three speakers, conversational, high energy, distinct voices. The intended mode. - **Verbal.** Well served in content, degraded in transcript by ASR corruption of every company name. - **Kinesthetic.** Well served. The discovery stack is concrete and startable same-day, with real prices named. - **Logical.** Moderately served. The argument chain is clear; the evidence supporting it does not withstand inspection, and the arithmetic is loose. - **Social.** Very strongly served — live audience Q&A, explicit invitations to come on stage, questions answered by name, genuine banter between speakers. - **Solitary.** Adequately served via transcript, though the digressions make it slower reading than the recap. - **Naturalistic.** Not applicable to this source — no natural-world classification content. - **Reading level.** Roughly grade 11–13 conceptually. Prose is casual and profane in places, but comprehension requires unglossed B2B fluency: ABM, MQL, MQA, XQA, PLG, RFP, intent data, first-party signals, demand waterfall, pipeline. ## 📊 Accessibility Assessment Mapped to WCAG 2.2 AA. 1. **Perceivable — text alternatives.** ⚠️ Auto-captions only, with every company name corrupted (ForgeX→"Forjax", Common Room→"Conorman room"). For a session whose value includes who to follow up with, that is a real loss. 2. **Perceivable — time-based media.** ⚠️ No audio description. Slide content, if any, is not described verbally; several references ("that play I mentioned") assume shared visual context. The blog recap partially compensates and is the more accessible artifact. 3. **Operable — keyboard.** ✅ Standard YouTube delivery. 4. **Operable — enough time.** ✅ Self-paced on replay. The live format had chat participation, which is a plus for some and a competing demand for others. 5. **Understandable — readable.** ⚠️ Dense unglossed acronym load, plus heavy verbal disfluency ("um," "like," self-interruptions) that makes the transcript harder to parse than the speech was to hear. 6. **Understandable — predictable.** ⚠️ Loosely structured. Speaker hand-offs are confused more than once ("Davis, that's all you first" / "Oh, I thought that was pass it to you"), and topics interleave with live chat questions. 7. **Robust — compatible.** ✅ Standard delivery; transcript has no headings or timestamps for navigation across ~5,900 words. **Priority fix:** corrected captions and a chaptered transcript. The recap already demonstrates that a structured text version of this content is achievable. ## 🌍 DEIB Assessment - **Cultural breadth.** Narrow. Entirely US B2B SaaS: dollar pricing, US vendor names, LinkedIn research, Forrester/Gartner as the benchmark of credibility, and a US-style enterprise bank as the complexity example. Nothing addresses how buying committees, hierarchy, or procurement norms differ internationally — a notable gap given the claim is about universal buying behavior. Will's opening invitation to say "where you're tuning in from" gestures at a global audience the content never accommodates. - **Inclusive language.** Mostly fine, with two snags. The janitor-versus-CEO joke recurs here and again uses a low-status occupation as shorthand for "irrelevant signal." And there is mild profanity ("freaking") that is unremarkable in context but worth noting for reuse in formal settings. Executive assistants are treated with genuine respect — "you don't make friends with your EAs" is framed as a mistake to correct, which is a small but real dignity for a frequently dismissed role. - **Neurodiversity support.** Weak. This is the least accessible document in the batch for anyone who struggles with unstructured input: heavy digression, three speakers interrupting and handing off unclearly, live chat interleaving, no chapters, no summary, and no visual anchor in the transcript. The recap exists and is far better on this axis — a reader who needs structure should be pointed there. - **Representation.** Limited. Three participants, all in similar commercial roles at similar firms, all presenting as US-based. No buyer, no end user, and no practitioner outside the ABM vendor/agency/advisory triangle. The audience does get real voice — Carlo's question is read and answered live, and attendees are invited on stage — which is more participatory than most vendor content. - **Psychological safety.** Notably good within the session. The host explicitly decentres himself ("this is not for me to hear my own voice"), invites disagreement, and flags in advance that there may be "healthy discourse" between guests. Speakers openly correct themselves mid-claim ($4,000 → $14,000), admit uncertainty ("this is some judgment call, so forgive me. There's not a perfect science"), and concede when the host reframes their point. Will offers his own conflicting statistic (6–10 vs 12–14) without defensiveness. That willingness to be visibly imprecise in public is the session's most trustworthy quality. ## ✅ Logical Validity **Strong reasoning** - **The 5% base rate is the session's best argument** and is genuinely load-bearing: if sales only reaches whoever the champion permits, and almost nobody runs marketing into pipeline, then most committee members are structurally unreachable. That is a real mechanism, not a slogan. - **The insurance anecdote is excellent evidence of a specific kind** — it demonstrates that blockers are often procedural checklists rather than persuadable opinions, and it is falsifiable and concrete. - **The gatekeeper/blocker distinction is precise and useful**: opinion versus no opinion, and it implies different tactics. - **Potter's refusal of the binary** ("it's not an either/or") complicates the vendor's own pitch and is the more defensible position. - **Cosby repeatedly marks his own uncertainty** — "some judgment call," "there's not a perfect science if you have zero data," "it takes 12 months" — which is unusual candour for content designed to sell. **Fallacies** 1. **Correlation presented as causation — with the confound volunteered and then ignored.** *Located:* the 30-vs-60-day claim, immediately followed by > [!quote] They're also not exploring other vendors or competitors. *Why it fails:* he then explicitly contrasts these with five-vendor RFPs and calls them "totally different." He has described **two different deal types**, not one type accelerated by marketing. A sole-source deal closing faster than a competitive RFP is unremarkable. The causal claim does not survive his own aside. *Aggravating:* the blog recap reproduces the statistic and **omits the confound entirely**, converting a hedged observation into a clean marketing proof point. *Fix:* segment by competitive context and report both cohorts. 2. **Survivorship bias in the taxonomy.** *Located:* the eight roles, derived from won deals. *Why it fails:* roles observed only in wins cannot explain losses. *Mitigating:* here he *does* compare against losses on the single-threading dimension, which is better than the recap implies — but the role list itself remains win-derived. 3. **Arithmetic inconsistency undermining the dataset.** *Located:* "338 sales cycles... 118 of those were closed one... the remainder 211," later "339 whatever sales cycles." *Why it matters:* 118 + 211 = 329. Three different totals appear. These are conversational approximations being presented as a research base. *Fix:* publish reconciled figures. 4. **Hasty generalization from one practitioner's book.** *Located:* throughout. One agency, one segment, three years, self-reported, generalized to B2B buying broadly. *Fix:* label as case study. 5. **Conflict of interest.** *Located:* structural — Common Room hosts and sells buying-group identification; both guests sell ABM services. Every supporting number comes from someone who profits if you believe it. 6. **Appeal to unverifiable authority.** *Located:* "some research from LinkedIn the 12 to 14 people." No citation. Compounded by Will offering a conflicting 6–10 and treating the discrepancy as confirmation of an upward trend rather than as two unreconciled numbers. 7. **Anecdote generalized to rule.** *Located:* the insurance story → "Just buy it regardless." *Why it overreaches:* one blocker was removed by one purchase in one segment. The underlying advice (find and remove procedural blockers) is sound; "buy it regardless" is not a general policy. **Tone of voice.** Energetic, informal, jokey, occasionally profane. Cosby is passionate to the point of interrupting himself and says so ("Sorry, I get really passionate about this"). Potter is measurably more careful and hedged — he is the one supplying caveats and complexity. Will hosts warmly and self-effacingly, deliberately keeping himself out of the spotlight and pushing questions to the audience. The register is peer-to-peer practitioner talk, assuming shared frustration with ABM tooling. Nobody is attacked; criticism targets vendors-in-general ("the big three") and practices, never named people. ## 🛡️ DARVO Analysis **Not applicable to this source.** No individual is accused of wrongdoing, so the pattern has nothing to attach to. - **Deny.** ⚪ Absent. The nearest analogue is the omission of the competitive-context confound from the downstream recap — but within *this* document the speaker states it voluntarily, which is the opposite of denial. - **Attack.** ⚪ Absent. Criticism lands on unnamed "big three tech vendors from 2020" and on practices (MQL treadmill, account scoring). Criticizing an anonymous vendor category is not an attack on a party, and no competitor is named. - **Reverse victim and offender.** ⚪ Absent. The trust-collapse narrative casts sales as wronged and marketing as the proximate cause, but it is a workflow diagnosis rather than blame reassignment following an accusation — and both functions are ultimately framed as failed by tooling, which the host's employer sells a remedy for. Do not tag this document with any `#darvo/*` value. ## 🎯 Conclusion **Bottom line.** The full session is more honest than its own recap. It supplies the eight-role map with sharper definitions, an excellent concrete demonstration that blockers are often procedural rather than persuadable, and a genuinely useful base rate — roughly 5% of organizations run marketing into open pipeline — that constitutes the real argument for marketing's role in a committee sale. It also volunteers, in one passing sentence, the confound that undoes its headline statistic: the fast deals had no competitors. Take the framework, the insurance lesson, and the 5%. Discard the 30-versus-60 figure, and note that the recap laundered it by dropping the caveat. **Next steps.** 1. Identify your own "insurance policy" — the cheap procedural artifact whose absence is quietly killing adoption. Ask a lost adopter directly; they will not volunteer it. 2. Use the 5% figure as the DevRel investment argument: sales meets only who the champion allows. 3. When citing either document, cite this one — it is the primary source and it contains the caveat. 4. Cross-link with the Josepha and Harbuz atoms: three sources now describing measurement that counts activity rather than the thing that matters. ## 🧱 Refactor Appendix ### Literature Split Pad #### L1 — Marketing reaches the committee members sales structurally cannot `#atom/literature/h1` > [!quote] Roughly 5% of them were running any kind of marketing in pipeline which functionally means that sales is only getting access to who their champions enable them to get access to. Claim: access asymmetry, not effort, is why committees go unreached. The strongest argument in the session and absent from the recap. #### L2 — Blockers are often procedural, not persuadable `#atom/literature/h1` > [!quote] you lose so many deals, you do so much work to then lose over a freaking $14,000 a year insurance policy. Just buy it regardless. Claim: the technical/implementation blocker frequently applies a checklist you never saw. The fix is operational (buy the artifact), not rhetorical (make a better case). #### L3 — The fast deals were different deals `#atom/literature/h2` `#fallacy/correlation-causation` > [!quote] They're also not exploring other vendors or competitors. Claim: the 30-day cohort was non-competitive; the 60-day cohort included five-vendor RFPs. Speaker calls them "totally different." This one line negates the causal reading the companion recap promotes. #### L4 — Gatekeepers hold no opinion; blockers do `#atom/literature/h2` > [!quote] Gatekeepers are different from blockers in my mind because they're people that are intended to protect time but they don't have an opinion. Claim: a precise distinction implying different tactics — befriend the gatekeeper, surface and answer the blocker. #### L5 — Blockers operate off-call `#atom/literature/h3` > [!quote] they aren't vocal on the call. They just sew seeds of doubt behind the scenes that you never get to actually directly address unless your champion's amazing and brings them to you. Claim: the most dangerous committee member is by definition invisible in your meetings — which is why champion quality determines whether you ever meet the objection. #### L6 — End users govern renewal, not acquisition `#atom/literature/h3` Claim: end users barely affect the first sale but determine "the second, third, fourth, seventh." The retention argument for post-sale investment. ### Atomic Split Pad - **Z1 — Access asymmetry `#atom/zettel/z1`.** In any gated relationship, one party can only reach whoever the intermediary permits. Broadcast channels reach the gated parties directly. This is the structural case for marketing, for community, and for public documentation. Links to `[[the-negative-space-question]]`. - **Z2 — Procedural blockers `#atom/zettel/z1`.** Many failures attributed to persuasion are actually unmet checklist requirements. They are cheap to fix, invisible until asked about, and never volunteered. Ask lost prospects what *rule* stopped them, not what they *thought*. - **Z3 — Confound volunteered, then discarded `#atom/zettel/z1`.** Speakers frequently supply the evidence against their own causal claim in an adjacent sentence, then continue. Reading for the aside rather than the assertion is a durable analytical habit — and downstream summaries reliably delete the aside. Links to `[[survivorship-derived-taxonomies]]`. - **Z4 — Opinion vs access `#atom/zettel/z1`.** Two kinds of obstacle: those who disagree and those who merely control passage. Conflating them wastes persuasion on people who need only courtesy. - **Z5 — Recap laundering `#atom/zettel/z1`.** A summary of a hedged source tends to strip the hedge, converting a qualified observation into a clean claim. Always read the primary when the statistic matters. Demonstrated concretely across these two documents. ### Fleeting Queue - `#atom/fleeting/f1` The recap dropped the confound. That's a clean, provable case study in how content marketing launders evidence — and I have both documents in the library to prove it. Strong Field Notes candidate. - `#atom/fleeting/f1` "What's our $14,000 insurance policy?" is a genuinely good question to put to any project with stalled enterprise adoption. WordPress version: SBOM? security contact? SOC 2 story for .org? - `#atom/fleeting/f1` 118 + 211 = 329, not 338. Three different totals in one talk. Nobody in the chat noticed, and the recap repeated the number. - `#atom/fleeting/f1` "Trust between buyers within a buying group is eroding" — raised by the host, never developed, and more interesting than anything else in the session. - `#atom/fleeting/f1` Four sources in this batch now on measurement-that-misses: Josepha (public toilets), Harbuz (funding recursion), and both halves of this ABM pair (80 points). The through-line is real. - `#atom/fleeting/f1` Zero mention of GDPR across both documents, in a method built entirely on identifying named individuals from behavioural signal. <!-- deep-read:sections end --> 📄 **Full text:** [[Full Document Contents/Articles/Accounts Don't Buy. People Do..md|Accounts Don't Buy. People Do. by YouTube]] ## Highlights - now get better deals. Crazy. But like, what marketing agency makes that purchase unless they it's a it's a real objection. Um, so someone finally told us, "We can't sign because you don't have insurance." Great. Let me go buy it. helpful. Sorry, I get really passionate about this because it's just you lose so many deals, you do so much work to then lose over a freaking $14,000 a year insurance policy. Just buy it regardless. Then you get blockers who are just like, I actually think this is the wrong thing for us to do. And they aren't vocal on the call. They just sew seeds of doubt behind the scenes that ([View Highlight](https://read.readwise.io/read/01kze813aza4r191gpmtve02hd)) - Note: L2 — The best concrete artifact in either document. They kept stalling at IT review; the actual reason was no cyber insurance; a $14k/yr policy removed the blocker. BLOCKERS ARE OFTEN PROCEDURAL, NOT PERSUADABLE — a checklist you never saw, not an argument you can win. The fix is operational (buy the artifact), not rhetorical. DevRel/OSPO translation: what's your $14k insurance policy? Missing SBOM, no security contact, unclear license, no SOC 2 story. Ask a stalled adopter what RULE stopped them, not what they thought. Also note the second half: blockers "aren't vocal on the call" — the most dangerous committee member is by definition invisible in your meetings. - Tags: go-to-market, buying-committee, atom/zettel/z1, atom/literature/h1 - sellers are already doing this. That data just never gets back to marketing. And again in the I've now I've had 339 whatever sales cycles which means I've had deep dive discussions with that many people about their specific programs. Roughly 5% of them were running any kind of marketing in pipeline which functionally means that sales is only getting access to who their champions enable them to get access to. Whereas marketing can actually just run ads. So like we can run ads that are case study ([View Highlight](https://read.readwise.io/read/01kze80ezj89as3tr1kz6gxv0m)) - Note: L1 — The strongest argument in the session, and it's absent from the recap. ACCESS ASYMMETRY: sales can only reach whoever the champion permits; only ~5% of orgs run any marketing into open pipeline; therefore most committee members are structurally unreachable. This is the defensible, non-vanity case for marketing/DevRel/community investment — broadcast channels reach the gated parties directly. Note also the loose arithmetic: "339 whatever" here vs 338 stated earlier vs 118+211=329. Three totals in one talk. - Tags: atom/zettel/z1, atom/literature/h1, abm, devrel-measurement - data, those that have actually where we've gotten more folks engaged in the buying committee earlier, our average sales cycle is 60 days. Those that have shown engagement earlier in marketing before they book the meeting have a 30day sales cycle. They're also not exploring other vendors or competitors. It's I just need to figure out if we can actually make this happen or I'm going to have to figure out how to use all your content for free to do this internally. Totally different sale than the well here's the RFP and you're going to be one of five vendors that we're ([View Highlight](https://read.readwise.io/read/01kze808s0bt1yw8srqtxz6yvn)) - Note: L3 — THE confound, volunteered and then discarded. "They're also not exploring other vendors or competitors... Totally different sale than the well here's the RFP and you're going to be one of five vendors." He has described TWO DIFFERENT DEAL TYPES, not one accelerated by marketing. A sole-source deal closing faster than a five-vendor RFP is unremarkable. CRITICAL CROSS-DOC FINDING: the Common Room blog recap (01kze5620hqprv2zdxwcycsev0) reproduces the 30-vs-60 statistic and OMITS this caveat entirely — converting a hedged observation into a clean marketing proof point. Read the primary when the statistic matters. - Tags: b2b-sales, atom/literature/h2, abm, fallacy/correlation-causation