// Earned presence
The machines trust sources, not claims
We measure which sources the engines actually cite in your category, then get you into them. Editorial only. No farms. No PBNs.
Offsite work is a black hole, and everyone knows it
A sponsorship here. A few Reddit threads. An affiliate pitch somebody said yes to. A retainer with a PR firm that reports placements rather than outcomes. None of it comes back with a read on which of those things, if any, changed what an engine says about you. So budget gets spent on the activities that are easiest to buy rather than the ones that matter, year after year.
The scale of it is measurable. AirOps counted 21,311 brand mentions across ChatGPT, Claude and Perplexity in October 2025 and found 85% came from domains the brand did not own. One scope note matters and usually gets dropped in the retelling: that study measured discovery and early vendor consideration, not the moment of purchase. It tells you where buyers first meet you, which is exactly the problem this service exists for, and it does not tell you what closes a deal.
The reason it stays a black hole is that citations are fragmented. An engine answering one question about you might pull from a dozen different domains, and no single placement moves that. Coverage across the sources that actually feed your money prompts beats any one prestigious win, which means the first job is finding out what those sources are.
Most authority programs aim at the wrong list
The standard approach starts with a media list: the publications somebody decided are prestigious in your category. It is a reasonable guess and it is usually wrong, because the engines are not reading the same list.
We know because we measured it. Running a live crypto brand's questions through every major engine and ranking the domains that came back, the single most cited source about that brand was a competing exchange's explainer page. Reddit, YouTube, aggregators, on-chain data and audit firms all outranked the trade press. Three or four of the top twenty were publications at all.
So we measure the list instead of assuming it
Engines attach their sources to their own answers. That means the ranked set of domains an engine trusts on your category is recoverable directly from what it says, and it is specific to your category rather than to the industry in general. That measurement is the first thing we do, and it is the thing a traditional PR process cannot produce.
Some of this is editorial link building in the plainest sense: earning coverage and a link on a site that already has standing in your category. We do that work, we pitch it on original data rather than on a template, and we buy nothing. It is one lane of several, and the measurement above is what decides how much of the budget it deserves.
From there each target gets classified, because the play is completely different for each kind. Aggregator and data listings are mostly forms and verification, high hit rate, no relationship required. Reference sites take a sourced correction through their own process. Community presence means participating honestly and disclosing it. Editorial needs a pitch, and original data is the pitch. Competitor-owned explainers usually cannot be moved at all, and we report those as context rather than quietly billing you to chase them.
Benchmark studies
The most citable thing you can publish is a fact that did not exist before you measured it. That is what a benchmark study is, and it is why the format earns citations for years rather than weeks.
We ran it on ourselves before selling it. Our audit of 146 crypto and fintech sites found that 95% are wide open to AI crawlers and still largely invisible in AI answers, which reframed the whole category conversation away from access. That is the shape of the thing we would build for you.
What we claim, and what we do not
Our own research on crypto and fintech brands found that third-party presence is associated with appearing in AI answers at all, and that once brand size is held constant it has no measurable relationship with how often engines cite your own site. So this service is sold on getting into the conversation. Anyone selling you links as a citation-rate guarantee is describing a relationship that did not survive the control.
Straight answers.
How do you decide which sites to target?
We measure them rather than nominating them. Engines attach their sources to their own answers, so the ranked list of domains an engine trusts on your category is recoverable directly from what it says. Every media list we have seen assembled by reputation disagreed with what the engines were actually citing, and the engines are the ones handing out the citations.
Is this link building?
Editorial link building is part of it, and it is the smaller part. Running this on a live crypto brand, only three or four of the twenty most cited sources were publications. The rest were aggregators, on-chain data, community threads, video, reference sites and audit firms. Traditional link building spends its whole budget on the smallest bucket on the list, which is why we run it as one lane inside a wider program rather than as the program.
Do you use link farms or PBNs?
No, and not only for the ethical reason. Bought networks are built to look like coverage to a crawler, and engines increasingly weight community and editorial signals that a network cannot fake. We also do not ghostwrite community posts or run undisclosed accounts. In a category where trust is the product, getting caught costs more than the placement was ever worth.
Will this get us cited more?
It is associated with being in the conversation at all, which is the strongest relationship we have managed to measure. In our own study of crypto and fintech brands, third-party presence predicted whether a brand appeared in AI answers, but not the rate at which engines cite the brand’s own site once brand size was held constant. We sell the first claim and refuse the second, because that is where the data actually landed.
What is a benchmark study?
Original research that creates a fact which did not exist until you measured it. That is why the format earns citations for years while a listicle earns them for weeks: engines cite facts, and a fact with one source has one place to point. We ran the program on ourselves before selling it, and the AI-Readiness Index is what came out.
See what the engines see.
Run the free scan and see your site the way the machines see it. No login. When you want the full picture, the Audit takes three weeks.