Solution · Startups

AI doesn’t care how old your domain is.

SEO rewarded incumbents — old domains, link history, brand authority. Things you don’t have. AI rewards specificity, real answers, and conviction. Things you actually do have. The playing field re-set, and almost nobody at your stage is competing yet.

The cold-start ledger

Day 1: zero domain authority, no backlinks, no brand recall. Day 30: cited in six AI shortlists, two of them on prompts your largest competitor still misses.

Median across our last 9 startup engagements · pre-seed to Series A

The 30-day playbook

From cold to cited. In four weeks.

One week, one outcome, one shippable artifact. Founder spends ~3 hours a week. We ship the rest. No 60-page deck, no 90-day “discovery phase”.

Week 101 / 04

Find the prompts that matter.

Founder

30-min call: who buys, what they ask, who you compete with.

Geology

Pull the 60–120 buyer prompts your category actually generates across all 5 AI platforms. Map where you stand on each.

Artifact

Prompt grid · 5 platforms × your top 60 prompts

Week 202 / 04

Mine your founder’s head.

Founder

Send us your sales call recordings, support thread exports, and the doc you keep meaning to write but haven’t.

Geology

Extract the question-and-answer pairs your conviction has actually formed on. Convert into specific long-tail content briefs.

Artifact

Brief library · 20–40 source-page briefs ready to ship

Week 303 / 04

Ship the source pages.

Founder

30 minutes per page reviewing the draft. Edits where it matters; ship-as-is where it doesn’t.

Geology

Draft, edit, and ship the source pages directly to your CMS. Schema, llms.txt, structured citation markup.

Artifact

Live pages · 8–14 specific, citation-friendly URLs

Week 404 / 04

Earn the off-site signal.

Founder

Approve the Reddit and community plays. Confirm tone. Don’t comment yourself.

Geology

Run the Reddit + community + small-creator program in your category. Real account history, mod rules respected.

Artifact

First citations · weekly tracking dashboard live

Scope split — who owns what

You ship the conviction. We ship the system.

The agency model breaks at this stage because it expects a marketing org you don’t have. Our model expects a founder. About three hours of your time a week, plus the calls you’d be on anyway.

You / your team
Geology
Conviction & POV
Yours. We don’t invent your viewpoint.
We extract it from your call recordings and Slack threads.
Source-page drafts
30-min review per page. Redline what matters.
Drafted, schema-tagged, citation-friendly, shipped to your CMS.
Reddit + community
Approve tone. Don’t comment yourself.
Real-account history, mod rules respected, citations earned.
Customer interviews
Yours — you’re the founder, you take the calls.
We mine the recordings for prompts and content seeds.
Tracking & weekly review
30-min weekly Loom or sync. Look at the numbers.
Live dashboard. Shortlist share, citation count, branded search, signups.
Paid ads
Yours. We don’t run paid.
Out of scope (and we don’t pretend otherwise).
One story, told straight

What this looked like for a Series-A devtools startup.

Two-founder team, ~$2M ARR, twelve months out of YC. Their category had three larger incumbents who owned the “best [tool] for [use case]” shortlist on every AI platform. Their own domain — six months old, ~40 backlinks — couldn’t rank for any of those queries on Google, let alone get cited.

The founders had spent a year talking to thousands of users. They knew exactly which workflows their tool handled better than the incumbents and which ones it didn’t. They had Notion docs. They had call recordings. None of it was anywhere AI could read.

We pulled their prompt set (87 buyer prompts), mined the call recordings and the founders’ Slack threads for the specific answers nobody else had, and shipped 14 source pages over weeks two and three. The Reddit work started in week four — three accounts with real history, helpful answers anchored to the source pages we’d just shipped.

By day 30 they were cited in 6 of the 87 prompts — six more than week zero. By day 90 they were cited in 41, and the largest incumbent was missing on two of them. The founders kept their same three-hour weekly cadence the entire time. Inbound demos from “heard you on ChatGPT” became their second-largest pipeline source.

Anonymous · Series-A devtools · 2026

Founder questions

What founders actually ask.

We have ~$0 marketing budget. Can we still afford this?
Maybe. We size the engagement to your stage — early-stage scopes are smaller and tighter than what we run for funded growth-stage companies. We also won’t take an engagement we don’t think will move pipeline for you in the first 60–90 days. If your category genuinely doesn’t have AI buyer behavior yet, we’ll tell you that on the kickoff call and you’ll save the money for when it does.
Should we just learn to do this ourselves?
If you have one person who can spend 15+ hours a week on it and is technical enough to ship schema, yes. Most early-stage teams don’t. The DIY route works if you’re willing to make GEO a major founder time commitment for 6 months. Hiring us is the “my time is more valuable on the product” trade — and we structure the engagement so it stays that way.
We just got funded. Should we also hire a full-time content / SEO person?
Eventually yes. Right now no. The first GEO hire makes sense once you have signal that the channel is working — typically 3–6 months in, when you have weekly citation data, repeatable plays, and a backlog of work. We’ve handed off to in-house teams several times; we’ll tell you when we think it’s the right move.
Our product is technical / niche / B2D. Does GEO matter for us?
Often more, not less. Technical buyers ask AI for tooling shortlists more than any other audience we’ve seen — Stack Overflow has been replaced by ChatGPT for a non-trivial slice of their research. Niche categories also have less competition for the AI shortlist, which means your conviction can move it.
What does this cost?
We size on the kickoff call against your stage and category, not off a published rate card. Early-stage engagements scope smaller than mid-market. We send pricing after the call, only if we both think it’s the right time.
Get started

From cold to cited. In 30 days.

Run a Live Audit or book a 30-minute founder call. We'll tell you up-front whether your category is ready and what the first 30 days would actually look like — pricing only if we both think it's the right time.