Ichnos Reviews and Details

Questions & Answers

1. What makes Ichnos unique?

   It reads the destinations rather than listing them. Anyone can produce a list of publications for a restricted industry — an AI will do it in seconds, and roughly a quarter of what it returns won't resolve. What nobody else does is fetch each one, read its published advertising or submissions policy, and record whether outside businesses may take part, quoting the sentence that says so.

   Of 7,178 places surveyed, 1,607 publish a documented route. The other 78% look identical in a list. Reading them is the product.

   The second thing is a finding: almost none of the available access is guest posting. It's advertising pages, press-release wires, business directories, event prospectuses and memberships. Anyone searching for "write for us" opportunities finds nothing and concludes the category is closed. It isn't closed — it's differently shaped, and that single misreading is why nobody else is here.

2. Why should a person choose Ichnos over its competitors?

   The comparison isn't really like for like. Media databases are built for pitching journalists — they assume you're allowed to advertise and just want earned coverage as well. Ichnos assumes the opposite: that the ordinary channels are closed to you entirely, which is the situation for cannabis, firearms, bail bonds, med spas, payday lending and ninety-odd other categories.

   Three practical differences. Every route carries the quoted sentence from that site's own page, so you can disagree with a classification rather than take our word for it. Access is sorted by what it actually is — what you can buy today versus what needs a relationship or an application — because those are different budgets and different quarters. And an industry below 18 confirmed-open routes isn't sold at all; below that it's a handful of guesses with a price on it.

   Pricing is also different in kind: one payment per industry, $497 to $2,997 by survey depth, with no seat licence and no annual contract.

3. How would you describe the primary audience of Ichnos?

   Marketing leads, practice owners and business-development staff at businesses in categories the major ad platforms refuse or restrict — cannabis and hemp, firearms, bail bonds, medical aesthetics, telehealth, payday lending, gambling, addiction treatment, and about ninety more.

   The specific person: someone whose ad account was rejected for what their business is rather than anything it did, who has a quarter to plan anyway, and who has been reading marketing advice written for companies the platforms are happy to take money from.

   Secondary: agencies serving those categories, who currently guess at channel and lose accounts when a placement is refused.

4. What's the story behind Ichnos?

   It started from a question with no good answer: where can a licensed dispensary advertise? Every result was an agency blog recycling the same six suggestions, none of which said whether the destination would actually accept anything from an outsider.

   The name is Greek — ἴχνος, a track or a trace. Ichnology is the science of reading tracks: establishing where something went from what it left behind, without ever seeing it. That's the method. You can't observe where a locked-out industry's audience gathers, but every industry leaves a trace — the private vocabulary it uses among itself. Follow the language and you find the room.

   The refusals came out of building it. It never posts anything, never marks a route open on its own authority, and never sells contact details — a route is a place, not a person. A product sold to people who cannot advertise, which itself worked by spamming, would be ridiculous.

5. Which are the primary technologies used for building Ichnos?

   Next.js and TypeScript on Vercel, Postgres via Supabase, Stripe for payments. Discovery and classification use large language models for the language layer only — never for the permission layer. Whether a destination accepts outside material is always established by fetching the page and reading it, never by asking a model what it thinks. That distinction is the whole product.