The Pre-Sale Engine
A way to package the audit so anyone can run it, not just the founders. One discovery call feeds three things at once: the qualification decision, the AI perception report, and a production-shaped demo account. Collect the data once, never build a throwaway.
Take the audit off the founders’ plate
Right now the audit is the wedge, and the people who can run it well are the people who built it. That works at design-partner volume. It doesn’t work when the goal is for anyone to sell, and it definitely doesn’t work at mid-market deal counts. The engine is how you make the motion repeatable: a script and a system, not a founder in the room every time.
One discovery call feeds three things at once: the qualification decision, the AI perception report, and a production-shaped demo account. You never collect the data twice, and you never build a throwaway. If they sign, the demo doesn’t get rebuilt, it gets activated.
That’s the whole efficiency argument. It’s what lets the audit run at mid-market volume instead of staying bespoke consulting every time, and it’s what lets a rep who didn’t invent the methodology still deliver it cleanly.
Five stages
Top of funnel
Two sources. A self-serve “lite audit” that gives a cold prospect a small, alarming taste of what AI says about them, fully automated. And outbound using that same lite audit as the opener. The free lite audit is the lead magnet. The full audit is the pre-sale.
Discovery call
The most important call. It qualifies AND captures every input the report and demo need. Designed so the prospect feels interviewed about their business, not processed through a form.
Full audit / report
Generated from the call. The wedge artifact. Shows them what AI actually says, benchmarked against the competitors they named, queried as the personas they named. It also surfaces the sources the models are leaning on to form that opinion, since that’s where the fix starts. The “12% vs 67%” slap-in-the-face moment lives here.
Demo account
Built in the same data structure as production from day one. Current AI perception scores, competitor benchmarks, the queries that matter to their buyers, the source map behind those answers, and a preview of the guidebook that fixes it. This is “see your fixed future,” and it’s the same screen they’d log into as a customer.
Proposal / close
The demo did the selling. This is scoping and signature.
Migration = activation
Because the demo was production-shaped, migration is just turning things on: live perception monitoring, AI-crawl tracking on their domain so they can see when models pick the content up, connecting the guidebook, expanding query volume. No rebuild. This is the scalability proof.
Every question pulls double duty
Each question qualifies the deal and feeds the report or the demo. That’s the design constraint.
- Have you noticed AI getting your brand wrong, or lost a deal you suspect went to a competitor an AI recommended? (need + urgency)
- Who owns AI visibility internally today, if anyone? (authority + champion vs blocker)
- Is there budget for brand monitoring or AI visibility, or would this be net-new spend? (budget reality)
- Is this a now-problem or a this-year problem? (timeline)
- Exact brand, product, category, and the positioning in one sentence. (what to query the models about)
- What do you want AI to say about you? (the gap analysis: desired vs actual)
- The highest-stakes questions a buyer asks an AI right before choosing you. “Best [category],” “X vs Y.” (these become the audit queries)
- What are you doing now for search visibility: in-house, agency, content engine?
- Anything specifically aimed at AI or LLM visibility yet? (almost always no, which exposes the gap)
- Who owns content and comms? (where Unusual plugs in and who signs off)
This set also silently scores their sophistication. A mature SEO operation grasps the GEO/AEO pitch instantly. A team with nothing is a longer education cycle.
- Who buys today? (direct / core ICP)
- Who do you want buying that mostly isn’t yet? (stretch)
- Dream logo? (ideal)
The audit should query the models as these personas. What a mid-market CFO hears when they ask ChatGPT about you is a different answer than what a developer hears. Persona definitions shape the queries, which makes the report sharper for mid-market instead of just prettier.
- Who do you lose to?
- Who do you get compared to?
- Who do you think AI recommends instead of you?
It hits every test
- It takes the founders out of the loop. The whole point is a motion anyone can run, so the people who invented the methodology stop being the bottleneck. The call is a script, the report and demo are generated, the rep carries it.
- Strategic, not executional. I’m describing the machine, not the next deliverable.
- Cost-aware. One call → three artifacts → zero rebuild is a direct answer to “how does the audit scale without staying high-touch?”
- Sharper for mid-market. Persona querying and competitor benchmarking make the report harder-hitting as deals get bigger, which is exactly the gap between the design-partner motion and a Whatnot CMO.
The blind spots, named before they get named
Posed as questions, not lectures. They double as the smartest questions in the room.
Running thousands of model queries per audit isn’t free. At mid-market volume, token cost is the real productization ceiling. Who eats that cost on a free or pre-sale audit, and where’s the line between the automated lite audit and the full one?
AI answers drift. A report from Tuesday can be stale by Friday. The product has to set that expectation, or the monitoring layer has to be the actual sold thing, not the one-time report.
The funnel needs a clear deal-size threshold where it goes self-serve through an agent vs human-AE-assisted. Naming that I’d want to define it shows I’m thinking about unit economics, not just motion.