Disclosures
The conflicts, stated plainly.
Answerable sells the work it measures. It sells audits, monitoring and fix work for AI visibility, and it also publishes the research about how invisible Australian businesses are to AI. That is a conflict of interest, and naming it is the point of this page. The scanner is Answerable's own instrument, so the scoring formula is published in full and the raw data behind the research is published under a CC BY 4.0 licence, which means a score can be checked rather than trusted. No vendor pays to appear anywhere on this site, and no payment is accepted to change a score.
The conflict, first
We sell the thing our research says you need
Answerable's research finds that most Australian businesses are not named by AI assistants when a buyer asks. Answerable also sells the work that tries to change that. Those two facts sit in the same firm, and anyone reading our numbers should read them knowing it.
We cannot remove that conflict, because removing it would mean either not measuring or not selling. What we can do is make the measurement checkable by someone who assumes we are motivated:
- The formula is published before the scores are. Seventy per cent of an engine score is a count of how often an engine named you across the questions it was asked. Twenty per cent is a recognition flag and ten per cent is sentiment, both graded by a model from the stored transcript. Those two are the only model-produced inputs in the pipeline, they are worth at most 30 of 100 points, and we say so rather than letting the number read as fully mechanical.
- The counted part is reproducible and the graded part is not. A third party holding the stored answers can recompute the 70 per cent term. They cannot reproduce the other 30. The honesty there is asymmetric and it is stated on the technology page in those terms.
- The research file is public. Every headline number in our 300-business study can be recomputed from data.csv with no access to us at all.
The clearest evidence that we publish against our own interest is the study's correlation between site readiness and per-engine visibility: a Pearson r of 0.116 across the 137 businesses with complete records from both engines. Site work is something we sell. That number says site work explains almost nothing about who gets recommended, and we published it anyway. The reasoning is on the editorial standards page.
The rules we hold
What Answerable does not do
-
No paid placement
Nothing on this site can be bought into. No ranking, list, glossary entry, industry page or example is sold, sponsored or exchanged. If a business or a tool is named here, it is named because the page needed it named.
-
No paying to move a score
No payment is accepted to change, raise, withhold or delete a score. Buying an audit buys the full set of stored answers and a prioritised fix plan. It does not buy a different number, and the number is produced by the published formula from the stored transcript either way.
-
No affiliate or referral income
Answerable takes no commission, affiliate fee or referral payment for any tool, platform or vendor named on this site. The only money the firm takes is the money on the how we make money page. If that ever changes, that page and this one change first, and the change is logged on the corrections page.
-
No partnership with any AI company
We are a paying API customer of two AI providers. We have no partnership, integration, affiliation or authorised access with OpenAI, Google, Anthropic, Perplexity or xAI, and we do not display their logos as endorsements. Two engines are queried on the deployed scanner, ChatGPT and Gemini, and which engines are switched on is a fact anyone can re-check at the scanner's own live configuration endpoint, linked from the technology page.
-
No guarantees about what an AI will say
No firm can promise what an AI assistant will say about you. Nothing in the scanner measures leads, revenue, traffic or rankings, so we do not claim the score predicts or causes any of them, we do not promise a score will improve, and we attach no timeframe or number to any improvement.
-
No publishing a client's numbers as marketing
Individual scan reports are unlisted, unguessable and set to noindex by design. A client's result does not become a statistic on this site unless they have agreed to it and the page says whose result it is.
Connected parties
Where we are connected to something we write about
The people behind Answerable also run a portfolio of Australian websites. Those sites are where the scanner gets tested before it is pointed at anyone else, which is useful and is also a conflict, because a firm marking its own homework can choose which homework to show.
The rule is a label, not an exclusion. Where a site connected to Answerable appears in published research, in an example, or as evidence anywhere on this site, it is labelled as connected to Answerable at the point where it appears. The label goes on whether the result is flattering or not. A connected site that scores badly is exactly the kind of result this cluster exists to publish, and today the clearest example is on the corrections page: both scanner defects logged there were found on sites connected to us, and both are written up with the businesses named and the measurements given.
The published study is a separate case. The 300 businesses in The State of AI Visibility in Australia 2026 have no relationship to us. It is the same engine described on the technology page, pointed at businesses with no connection to the firm, and the raw file lets anyone check who is in it.
AI Discovery is us. AI Discovery is the AI visibility practice within Answerable, not a third party, not an independent reviewer and not a second opinion. Where the name appears on this site or in a footer, it is the same organisation.
Upkeep
What would change this page
A disclosures page is only worth reading if it is maintained, and the failure mode is a page written once and left to go quietly out of date. So the commitments are specific enough to break.
If Answerable ever takes an affiliate fee, accepts a placement, enters a partnership with an AI provider, or starts publishing a connected site's result without a label, this page changes on the same day and the change is recorded with a date on the corrections page. If you find something on this site that contradicts what is written here, that is a correction and we want to hear about it: [email protected].
The related reading is the editorial standards, which cover what gets published and how, how we make money, which is the money side of the same question, and about Answerable, which is who is making the claim.
Keep reading
The rest of the trust cluster
Trust
Editorial standards
How we decide what to publish, and how the research sample is skewed.
Read more →Trust
Corrections
How to report an error, and the dated log of the ones we have made.
Read more →Trust
How we make money
Four prices in Australian dollars, and what no one can buy.
Read more →The firm
About Answerable
Who we are, where we operate, and our own score of 29 out of 100.
Read more →See what AI says about your business.
Start with a free scan. It is the fastest way to find out whether AI recommends you or your competitor, and you keep the report.