The method · documented in public
The Moncrieff Method
The Moncrieff Method is RankOps' five-step system for making a local business findable by Google and quotable by AI engines: audit the signals, fix the entity, publish the answers, build the citations, and track what AI actually says. It is documented publicly because the proof is live — 241 neighborhood pages across 25 North Carolina cities you can click through right now.
Is this SEO, or is it something else?
It is SEO, done for a search landscape where the answer is increasingly written rather than listed. Steps one to four are technical SEO, structured data, content and local citation work any competent agency would recognise. What differs is that they are sequenced and marked up so an AI engine can quote the result — and that step five measures whether it did.
Generative engine optimisation is an ingredient in this method, not the product. A business owner does not want GEO. They want the phone to ring, whether the customer found them in a blue link or in an answer ChatGPT wrote.
Why does the order of the steps matter?
Because each step is worthless without the one before it. Publishing great answers on a site an engine cannot identify wastes them. Building citations for a business whose name and address disagree across the web actively confuses the engines you are trying to convince. Entity first, answers second, citations third.
Most of what looks like an AI visibility problem is an identity problem wearing a costume. That is why step two exists before step three, and why skipping it is the most common reason good content never gets quoted.
Step 1 — Audit the signals
Read the site the way an AI reader does, not the way a person does: 16 weighted checks across findability, citation signals and local discoverability. The scoring is deterministic and repeatable, so the same site scores the same twice and a change in score means a change in the site.
It is the same scorer behind the free score check and the Site Inspection — the free version tells you the number, the paid version names every failing signal and ranks the fixes by impact.
What this catches: one real-estate site was serving a noindex tag that hid it from Google entirely, alongside seven other failures. Nobody had noticed, and the site was otherwise working fine. Read what that costs →
Step 2 — Fix the entity
Make the business a specific, verifiable thing rather than a name on a page: consistent name, address and phone everywhere it appears, business schema describing what it actually does, and an owner an engine can identify as a real person. Engines do not recommend businesses they cannot confidently identify.
This is the step people skip, because nothing about it looks like marketing. It is also the step that decides whether anything after it works. An engine deciding who to name is doing a confidence calculation, and inconsistent identity is what loses it.
Done to ourselves first: RankOps' own entity record is public and machine-checkable — the canonical facts, and one canonical identity referenced across every property we own.
Step 3 — Publish the answers
Write the questions customers actually ask, and answer each one in the first 40 to 60 words, marked up with FAQ and Speakable schema. That opening passage is what an engine lifts and quotes. A page that buries its answer under three paragraphs of preamble has nothing quotable in it.
The phrasing matters as much as the content. "Who is a good mechanic in Kannapolis" is a question a person asks an assistant; a services list is not an answer to it. Every answer page on this site is built the same way — read them and check the first sentence of each.
Step 4 — Build the citations
Get the directories and data sources an engine cross-checks to agree about who and where you are, then add genuinely local pages so recommendations can land block by block rather than city-wide. Corroboration is what turns a claim on your own site into something an engine will repeat.
Neighborhood-level content is the part most agencies stop short of, because it is slow. It is also what decides whether you are recommended in the suburb where the customer actually lives.
At scale: 241 live neighborhood pages across 25 NC cities, and 100 more on buy-sell.land — all built with this step, all clickable.
Step 5 — Track what AI says
Run the real questions on a schedule and record the answer: cited or not, and if not, who was instead. This is the step that turns visibility from a hope into a number, and it is the only way to know whether steps one to four worked for a particular business in a particular market.
Tracking also keeps everyone honest. AI answers shift as models update, so a single screenshot proves very little; a dated series proves a direction. That is why every number on this site links to a case study with a date on it.
Run on ourselves too: the dated log of what AI actually says about RankOps, including the runs where we were not cited.
How long does it take to see movement?
Search engines and AI assistants typically re-index new content within 30 to 60 days, which is when changes become measurable. Triple B's Mechanics went from zero AI citations to nine in 45 days. Results vary by market, competition and starting point, and AI citations can never be guaranteed by anyone.
Every figure above is from a named client with a dated write-up, not a claim on a slide. All four case studies →
Questions
What is the Moncrieff Method?
The Moncrieff Method is RankOps' five-step system for making a local business findable by Google and quotable by AI engines: audit the signals, fix the entity, publish the answers, build the citations, and track what AI actually says. It combines technical SEO, content optimisation, structured data, entity optimisation and AI citation work into one sequence, because the same signals feed both.
Is this SEO or is it something else?
It is SEO, done for a search landscape where the answer is increasingly written rather than listed. Steps one to four are technical SEO, structured data, content and local citation work that any competent agency would recognise. What is different is that they are sequenced and marked up so an AI engine can quote the result, and that step five measures whether it did.
Why does the order of the steps matter?
Because each step is worthless without the one before it. Publishing great answers on a site an engine cannot identify wastes them. Building citations for a business whose name and address disagree across the web actively confuses the engines you are trying to convince. Entity first, answers second, citations third.
How long does it take to see movement?
Search engines and AI assistants typically re-index new content within 30 to 60 days, which is when changes become measurable. Triple B's Mechanics went from zero AI citations to nine in 45 days. Results vary by market, competition and starting point, and AI citations can never be guaranteed by anyone.
See where your business stands today.
Two minutes, no call. Or go straight to the $99 Site Inspection.