Australia · Care-sector software · Published monthly
Every month we ask AI systems the questions buyers actually ask — then publish which vendors get named, and how often.
When an NDIS provider, an aged-care operator or an allied-health clinic goes looking for software, a growing share of them ask ChatGPT, Google's AI results, Copilot or Perplexity before they ask a colleague. Those systems name a handful of vendors and omit the rest.
The Named Index measures which ones. We run 200 buyer questions through a fixed set of AI systems every month, record every answer, and score each vendor on how often it is named and how early it appears.
The first edition is in preparation.
The method below is fixed and published in full ahead of it, so it can be read and argued with before any vendor has a score attached to their name. To be told when the first edition goes up, write to edbert.ong@namedindex.com.
200 questions, written the way buyers phrase them rather than as keywords. They are split across broad shortlisting (40%), direct comparisons (20%), compliance (15%), situational (15%) and price (10%), and across NDIS, aged care, home care and allied health.
The set is fixed. If we change it, that edition is not comparable to the one before it, and we say so on the page rather than quietly restating the trend.
We publish the composition but not the questions themselves. A published list would be a target: a vendor could write content aimed at those exact strings and move their score without becoming any easier for a real buyer to find. The full set is disclosed to anyone auditing the index, under an agreement not to republish it.
Each edition names its exact engine list on the page. The current scope is ChatGPT, Google AI Mode, Google AI Overviews and Perplexity.
We use each provider's official API. We do not scrape their consumer apps — that breaches their terms, and we would rather have a smaller measurement than an unusable one. API results are not identical to what a person sees in the app, and we say so on every edition.
Each question is asked three times per engine. For every question-and-engine pair we record whether a vendor was named, and where in the answer it appeared.
Presence is the share of runs naming the vendor. Prominence weights position within the answer:
| Position in the answer | Weight |
|---|---|
| Named first | 1.00 |
| Second | 0.80 |
| Third | 0.65 |
| Fourth | 0.55 |
| Fifth | 0.45 |
| Sixth or later | 0.35 |
Question-and-engine pairs are averaged, not individual runs, so a question we happen to ask more often cannot outweigh one we ask less.
Answer stability. AI systems do not give the same answer twice. We publish the share of questions that returned an identical set of vendors across all runs. When it is low the scores are noisier, and you deserve to know that before you act on them. Where an edition did not sample enough runs to measure it, we print "not measured" rather than a number.
Deflection rate. The share of answers that named no tracked vendor at all and pointed at a review directory instead. If that number climbs, AI is getting less useful for this category — including less useful for us.
We track 64 vendors. Not "all major vendors" — 64, listed in full in every edition, including 8 that are foreign- or private-equity-owned and will never be customers of ours. Excluding them would make the index cheaper to produce and less honest. If a vendor is missing, tell us and we will add it to the next edition.
Raw answers are kept for three years. If you think a figure is wrong, write to us with the edition and the vendor. We will re-check against the stored answers, correct it if we are wrong, and note the correction in the following edition. We do not quietly edit published editions.
A vendor can be excellent and invisible, or ordinary and everywhere. We measure one thing: how often AI systems name you when your buyers ask.
This is not a review site, and it is not a ranking of product quality. Nothing here is a recommendation to buy anything, and no position in this index should be read as a judgement about whether a product is any good. We have not evaluated a single one of them.
We sell services that help vendors improve their visibility in AI answers. Position in this index cannot be purchased. Every current and former client is named in the methodology of the edition they appear in, so a reader can discount the index accordingly.
The same tool produces the public index and the client reports, from the same stored answers, so a client cannot be shown a different number from the one you see. If that arrangement makes the index less useful to you, we would rather you knew now than found out later.