register: human · the front door

The Recommendation Audit.

Dual Cognition is an AI creative agency that builds campaigns to make your brand the one AI recommends. The Recommendation Audit is the front door of every campaign: a measured, live diagnostic of how AI models read and recommend your brand.

It runs $10K to 15K and takes two to three weeks. You get 7 measured modules inside a 10-section report, and every number in it carries its run id and date.

register: machineproduct spec

product : Recommendation Audit
price : $10K to 15K
duration : 2 to 3 weeks
modules : 7 · all measured
report : 10 sections + the campaign brief
free_tier : Recommendation Check · human-fulfilled

what we measure

Three modules of ours. Four powered by Alien Eyes. All seven measured.

The audit isn't an opinion. Each module returns numbers from live runs against the major AI models and your real surfaces. The three DC-unique modules read meaning. The four Alien Eyes modules read machinery, and we name our sources.

register: machinemodule ledger · 7 measured modules
01 the machine mirror live llm visibility · your cold ai recommendation rate, measured across the major models dc · unique
02 the entity fragmentation map how consistently every surface describes you dc · unique
03 the proof gap which claims have agent-discoverable evidence dc · unique
04 crawlability can agents reach every page that matters powered by alien eyes
05 structured data is your schema present, valid, and matching the visible page powered by alien eyes
06 parseability does your meaning survive being read without the design powered by alien eyes
07 performance does the site answer fast enough for an agent's patience powered by alien eyes

how it runs

Four moves, two to three weeks.

No black box. You can watch every step land, and every number is born with its run id attached.

week 0

Kickoff

Access and scope. You hand us the surfaces, and together we agree on the claim list and what counts as yours.

week 1

Measurement

Live runs. Cold queries against the major models, every surface read, every claim traced to its evidence.

week 2

Analysis

The runs become the 10-section report and the campaign brief: what to fix, in what order, and the number each fix should move.

final session

Walkthrough

We read the report together, page by page, finding by finding. You leave knowing exactly what AI says about you today.

the output shape

The output isn't a grade. It's numbers you can act on.

Most sample reports are airbrushed: rounded numbers, happy charts, no provenance. Ours leads with the number exactly as the run returned it. The one on the right is from a live engagement, and it stings, which is exactly why it's useful.

If your number is zero, you'll hear it from us first.

superhired · recommendation audita real measurement from a real engagement
ai recommendation rate · cold0%0 of 135 cold queries · measured live · run initial-v8 · 2026-05-22
entity consistency69.6/100 · measured · 15 surfaces · run initial-v8
cold = a buyer who has never heard of the brand. entity consistency is the leading number; the cold rate is the lagging one it pulls up.
Your report holds numbers shaped like these, about you, each with its run id and date. No props, no fabricated charts.

the smaller door

The free Recommendation Check.

It answers one question, plainly: is your brand visible to AI at all? A verdict, not a judgment. A person runs it, a person writes it, and it comes back to you in about two business days. If the verdict says you're fine, we'll tell you that too.

human-fulfilled · returned in about two business days

the obvious question

Why not just hire an SEO firm?

It's the question every buyer should ask. Here's the honest answer.

an seo firm

Optimizes pages.

Tunes your pages to rank in a list of links. Real craft, and still worth having. But it treats your brand as pages to be ranked, not as an entity a machine needs to understand.

a brand studio

Makes the work.

Makes the work people love: identity, story, film. Essential, and half the job. Then the machine reads nine beautiful, slightly different things and can't tell they're one brand.

dual cognition

Does the work, then makes it legible.

We make the work people love, and we make the machine read it as one brand. One message on every surface, measured before and after. That's the campaign, and the audit is its front door.

Know what AI says before your customers ask.

Or just start a conversation.