The Factory ships brand, landing page, audio pitch, financial model, and outreach kit for every idea. None of that ships customers. You do. This page is the part where we say what we actually believe about that.
The Factory is an autonomous studio that produces shippable starter kits for AI-business ideas at a rate of about 10 per day. As of this writing it has produced 822 idea kits, all of which now have operator-voice landing pages with no skeleton-broken content. Across that catalog the median idea has a brand, a one-page landing, an audio pitch, a Fermi-math financial model, and a draft outreach pack. The median idea has zero customers, zero validated demand signal, and zero revenue. One idea (Counsel AI) has graduated to its own domain at intakecounsel.com, a roughly 0.4 percent graduation rate in the studio's first six weeks. That is venture-portfolio-equivalent for unfunded operator launches.
That's not the Factory failing. That's the Factory working as intended. The job of the studio is to take a fuzzy market signal and make it concrete enough that a buyer can decide in 10 minutes whether the idea is worth their attention. The job of the buyer is to do everything that comes after.
What the dossier gives you: a head start of about 40 hours of strategic work that you'd otherwise do yourself. ICP, MVP scope, GTM channel, outreach pack, financial model.
What the dossier doesn't give you: taste, judgment, customer conversations, distribution, retention. None of that can be bought for $5 or $99 or $99,000 from anyone.
An idea is a hypothesis. A business is a result.
The catalog gives you a structured hypothesis. Whether it becomes a business depends on what you do in the next 90 days. We've watched smart people turn weak hypotheses into real businesses by sheer distribution work, and we've watched smart people sit on excellent hypotheses for two years and never talk to a single customer.
Distribution is harder than building.
Roughly 90% of the calls we get from buyers who unlocked a dossier are about implementation. The implementation is mostly tractable. The harder question, the one we ask first in any operator engagement, is: do you have access to 50 buyers in this ICP that you can talk to in the next 30 days? If the answer is no, the idea is worse than it looked on the page, regardless of the dossier quality.
Probability of meaningful success is in the 5 to 25% range for most ideas.
Each idea ships with a Fermi-style probability estimate (search "probability_of_meaningful_success" in the financial model). These numbers are derived heuristically. Read them as "the chance this idea will reach $144K ARR in year 1, given an operator with normal distribution access and reasonable execution." The number is rarely above 30%. When it's above 30%, it's because the ICP is in a category Wes already sells to.
No idea in this catalog has live customer revenue.
One product (Counsel AI / law-firm intake) has graduated to its own domain at intakecounsel.com and is being marketed (see graduated list). The rest are starter kits. We won't fabricate testimonials, results, or revenue figures. If you see a number on this site, it's either a stated estimate or it's pulled from public benchmarks.
The Factory is most useful for one of these buyers:
If you're hoping for a turnkey business that runs itself, this isn't that. You'll spend $5 and feel disappointed because the dossier won't tell you who to call. If you don't have access to a clear ICP yet, the dossier is less useful than three weeks of customer conversations.
If you want guaranteed revenue, you want an acquisition (a real company with real customers), not a starter kit. The two are different categories of asset.
If you want us to validate the idea for you in advance, that's the Phase A engagement on the operator partnership page. It's $2,500 and it explicitly includes the validation calls.
If you unlock a dossier and it leads to one of the following, that's a win:
If you unlock a dossier and the only outcome is that you read it once and then shelf it, that's a normal outcome too. The dossier is shareable, you can come back to it, and most ideas surface their value on the second or third read.
Five questions we get from buyers who are pretty sure this is too good to be true. The answers are the honest ones.
"Isn't this just AI slop dressed up?"
It would be, if we generated the entire dossier with one LLM call against a template. We don't. Each idea starts with a market-signal mining pass (Reddit, HN, niche communities, our own customer transcripts), then a buyer skeptic memo where we name a specific operator and write their objections in their voice, then an Adoptability scorecard against ten weighted axes, then a hand-edited elevator pitch, and finally a financial model with editable Fermi assumptions. The catalog landing pages now use LLM bulk generation (iter 58 of the autonomous studio), but the inputs to that generation are hand-written per-product dossiers, not generic prompts. The difference is whether the LLM is inventing or condensing. Ours condenses.
"How is this different from asking ChatGPT for a business plan?"
ChatGPT gives you a plausible business plan in 30 seconds. The Factory gives you a plan grounded in market signal we already scraped, scored against 822 alternative ideas, with a Fermi probability estimate that uses real category benchmarks (Mailchimp ARPU for nurture tools, Clari pricing for forecasting tools, EagleView per-report cost for roofing tools). The numbers are not from a vacuum. You can still get a plausible plan from ChatGPT for free. You cannot get the comparative scoring without doing the work ourselves.
"Why $5? What's the catch?"
$5 is the unlock price, not the business. The economics work like this: $5 unlock to read the dossier, $99 to commission a specific buildable artifact (custom outreach pack, financial model with your inputs, slide deck), $2,500 for a Phase A partnership where we run customer validation calls with you, $15-25K for a Phase B build where we ship the MVP. Most buyers stop at $5 and we are fine with that. The unit economics work at $5 because we produce 10 dossiers per day at roughly $0.50 LLM cost per idea. The catch is that $5 buys you the strategy work, not the customers or the build.
"What if I buy a dossier and the idea is actually bad?"
Every dossier ships with a probability_of_meaningful_success estimate in the financial model (typically 5-25%). If you buy a dossier and decide the idea is bad before you ship anything, we refund the $5 with no questions. If you ship something and it does not work, that is the operator risk we cannot refund. We try to label the risk honestly in the dossier so you can make the call before you spend three months on it.
"How do you know the dossiers are not stale?"
The catalog runs on a 24/7 cron schedule. Adoptability scores rescore every four hours against current market signal. Em-dash style invariants get swept every 15 minutes (yes, that is real). Content drift in any of the 627 product pages gets caught and corrected within an hour. We log every iteration of the autonomous studio in the changelog at /factory/changelog/. If the dossier you bought two months ago has stale data, it has already been refreshed and you can re-download.
You do not have to take our word for it. /factory/cron-status/ shows the 124 cron jobs running with live timestamps. /factory/quality-report/ shows the audit results (0 hard fake-proof findings, 0 broken taglines, 73+ health endpoints passing) refreshed every 30 minutes. /factory/api-docs/ documents the 5 read endpoints (state, adoptability scores, health, healthz, dossier teasers) you can poll directly. These are the surfaces a buyer can check before unlocking anything.
If you find an idea that resonates and you ship something based on it, we'd love to hear about it, whether the result was good, bad, or mid. We use that feedback to retune the Factory's scoring rubric. Email wes@wishdeal.com with the slug and a few sentences. No reply expected.
If you spot something genuinely wrong (a market that doesn't exist, an ICP that's been wrong for two years, a financial model with a bug) tell us. We update the catalog and the model improves for everyone.
If you read this page and want the operator-honest details on three things that often come up: