A Cold DM Wanted $199. I Read 22,000 Words Instead.
23 posts, 100 notes, and 6 contradictions I could check for free.The archive says more than the pitch does. | Generated with ChatGPTA cold DM landed in my Substack inbox with a specific compliment and a $199 offer.$199 for a roadmap, or $799 if I wanted them running the account for me. Halfway…
23 posts, 100 notes, and 6 contradictions I could check for free.The archive says more than the pitch does. | Generated with ChatGPTA cold DM landed in my Substack inbox with a specific compliment and a $199 offer.$199 for a roadmap, or $799 if I wanted them running the account for me. Halfway through the conversation, a third price appeared between the two.The DM had opened with a compliment on my “workshop” branding. They had read the publication long enough to name the thing I care most about. That worked on me. I do like my workshop branding.TL;DR: I could not verify a single number in the pitch on its own, but their whole archive was sitting in public and I could check the pitch against that. So I read it instead of buying: 23 posts and about a hundred notes, roughly 22,000 words of it, with an AI doing the reading in one evening.The rest was standard. Six Substack bestsellers is a credential, and $15K to $96K is a range wide enough that any number inside it counts as a hit. Ten million downloads came with no denominator, and nobody said downloads of what, or spread across how many years. Then the closer: exactly one spot open.The offer itself was a strategic-advising roadmap. Two prices arrived first: $199 for the plan and $799 for full account handling. Then a $350 “high tier” showed up between them, mid-chat, without me asking, and I read the new rung as a test of what I would pay.The same chat told me there are no guaranteed numbers or timelines. A few messages later, it gave me a 45-day timeline to $7K a month.The textbook answer here is to archive the DM and get on with my evening. I didn’t, for two reasons. The claims were specific enough to be checkable, which is unusual. And checking them has become cheap in a way it wasn’t two years ago, which is the only reason any of this was worth an evening.The archive didn’t agree with the pitchSubstack archives are public. All I needed was their post list and a way to read it quickly.Pulling one takes a few minutes. Every publication has an archive page that lists its posts in order, and the free ones open without a login. I collected the ones that looked like case studies or income reports and saved the text into a single file.Then I gave that file to an AI with the pitch pasted at the top, and asked one question rather than asking for a summary: could the posts and the pitch both be true at the same time?It came back with a list of places where they could not. I opened the post behind each one before I believed it, because not taking a claim on trust was the entire point of the exercise. That is the habit I picked up when I stopped using AI as a search engine.The posts kept contradicting each other:Their paid subscriber count hadn’t moved since December.One post claimed five figures a month. Another disclosed lifetime product sales that didn’t add up to one of those months.The bestseller-client count doubled between two posts five days apart.The one-spot-left line appeared in every post that month.Their own publication grows about a subscriber a day. The case study they sell is 33x subscriber growth in eleven months.No nameable client anywhere in the archive. Every case study under NDA.Their disclosed paid line since December, against a claim of five figures a month. | Generated with ClaudeThe chat had its own version of this. Clients earning more than $50K a month, and then, a few messages later, the actual disclosure: two founders paying $1,200 a month.They have also published their own account of how a cold DM turns into a paid call. The cold DM is the business, and they had written up how it works. After that, I could see the next move in the chat before it arrived.It is all public. Apparently nobody reads it, including the people being pitched.When I declined, they offered a 100% refund guarantee. That is the takeaway close: the move where the seller takes on all the risk, so saying yes costs nothing. It is the last thing in the script, and by that point I had stopped reading the chat and started taking notes on it.About 30% of the archive is worth havingRoughly 70% is the same handful of frameworks restated across posts, which is what weekly publishing does to anyone. The other 30% is real methodology, mechanical enough that I could implement it without asking anyone.The profile page is a funnel. The bio line, my pinned note, the description, and whatever order a stranger happens to meet them in are all doing the same job. Mine is selling for me whether or not I meant it to, the way one embed ended up working across two platforms at once. I filled mine in on day one and never looked at it again.The offer ladder. Free at the bottom, expensive at the top, and at least one rung between them. I would not jump from zero to $199 on the strength of a direct message, and I doubt many people do.My own ladder has a large hole in the middle. I could only see it once I had a shape to compare it against.My ladder next to theirs, with the middle rung missing. | Generated with ClaudeThe anchor essay. One substantial piece a week, and everything else that week points back at it. Notes and comments feed it.I had been running notes and essays as two separate streams. Now the week’s notes all point at one essay, and of the four changes on this list, that is the one I would redo first.A discovery mechanism for products after launch. Launch day gave me one spike, then nothing. The weeks after needed a standing route in, a nav tab, and cross-links from every essay that touches the same subject.All four came out of posts they gave away. The $199 adds the order to do them in, and a person to tell me which of the four matters this month rather than next.Ninety days, and most of it is repair workSo I built the roadmap myself, mapped against my actual numbers. It took one sitting. It cost nothing, and I know why each line is on it.The result is unglamorous. Most of it is repair work on things I have already published: my profile page, the nav, the welcome email, and internal links between essays that should have been linked from the start. After that, one anchor essay a week with everything else pointed at it, and then the missing rung on the ladder, somewhere between free and my paid product.None of it is a secret.The ninety days are not evenly split. The first stretch is cleanup with nothing to show for it, because relinking old essays and rewriting a welcome email produces no new posts. The anchor-essay rhythm is the only item that changes what I publish rather than how it is arranged, and the middle rung is the only part of the plan I would call a bet.That middle rung is still missing, and it is the item I keep pushing down the list. The only new thing on the plan is Reel Pipeline Studio, the tool behind the short history videos I post, and I’m aiming to launch it in October.I have made this trade before. Kit wanted $39 a month to send five emails over two weeks, so I spent four days building the drip tool myself, and it has cost me a hosting bill ever since, currently around $4 a month on Fly. That was the same decision at a smaller scale: one evening of reading, no code, and I still ended up doing the work myself.Small numbers fit in my headHere are mine. They are small enough that I can check them. Month five: 257 subscribers, and sixteen Gumroad orders since June, nine of them at $0, $265 in total.Three months ago that same list read 74 subscribers and one paying reader.Month two next to month five, the only numbers I can verify myself. | Generated with ClaudeSixteen orders is short enough to recite from memory. I know roughly when each one arrived and which post it followed, and at this size that is easier than opening a dashboard. When I can trace a sale back to a specific essay, I write more of that essay, and when I can’t, I assume it came from somewhere I don’t control yet.The 74 is the line I look at most. Three months of work sits between that row and the current one, and every post from those months is still in my own archive, in the open, exactly as theirs was. Anyone who wants to do to me what I did to them can, and they will find a small publication that grows slowly and says so.The goal I am working toward is an honest $500 a month by November. I can name where every dollar of the $265 came from, which is why $500 is a number I can plan for. When one of these numbers moves, I notice, and I usually know why.I can’t picture tens of thousands of subscribers, let alone check them, and I have no way to check a $96K launch figure in someone else’s DM or to tell whether it ever moved.They may well be better at selling than I amTheir disclosed lifetime sales are well ahead of mine. That is what made the five-figure-month claim odd, because the smaller number was the believable one and they published it themselves.They built something that turns cold DMs into paid conversations, then wrote up the method. It worked well enough to reach someone who wasn’t looking for it. That’s a real skill, and I don’t have it yet.I happened to have the habit of pointing an AI at an archive and asking it one question, which is what I built for my own back catalog. That was my edge, and it isn’t much of one.What I got from the consultant was a live demonstration of the funnel, run on me. It opened with a compliment and closed with a refund guarantee after I had already said no, and every rung in between was visible from the archive I had just read.Drop a comment with a line from a pitch you still remember: the compliment that was a little too specific, or the number with nothing behind it. I want to see which lines are still doing the work.External SourcesSubstack — publishing platform with public post archivesGumroad — storefront behind the $265 in salesClaude — the AI that read the archiveI build software products solo and write about the calls that keep that sustainable: which costs are worth paying, and which ones I would rather work out myself. If that is useful, it is a free email series, Good-Enough Engineering.This story is published on Generative AI. Connect with us on LinkedIn and follow Zeniteq to stay in the loop with the latest AI stories.Subscribe to our newsletter and YouTube channel to stay updated with the latest news and updates on generative AI. Let’s shape the future of AI together!A Cold DM Wanted $199. I Read 22,000 Words Instead. was originally published in Generative AI on Medium, where people are continuing the conversation by highlighting and responding to this story.Source: Generative AI Pub — Published — Category: Image AI