The AI Answer Share Report
2026-Q3 · Perplexity Sonar · published by InterCore Technologies ·
The AI Answer Share Report measures which websites AI search engines cite, and which law firms they name, when someone asks which lawyer to hire. In 2026-Q3 we put 540 questions to Perplexity Sonar — six practice areas across 30 US metros in 23 states — and ran each question 3 times, for 1,620 total responses and 18,900 source citations. We report two separate things: Answer Share, whether a firm is named in the answer text, and Citation Share, whether a domain is cited as a source beneath it. A firm can be named without its website being cited, and that turns out to be common.
Read the full methodology — including how firm names are extracted, how sources are classified, and the seven limitations we know about.
What we found
When Perplexity answers a question about hiring a lawyer, the first source it cites is a legal directory.
share of answers whose first-listed source is a directory. 1,254 of 1,612. Based on 540 prompts across 1,620 runs, measured .
Legal directories account for at least half of every source Perplexity cites on lawyer-hiring questions.
share of all citations going to known legal directories. 9,795 of 18,900. Based on 540 prompts across 1,620 runs, measured .
This is a floor, not a point estimate. A floor, not a point estimate: 36.0% of citation volume sits on domains not yet classified, some of which are further directories. The true directory share is higher.
Ten domains absorb nearly half of all citations, out of 1,749 cited.
share of citations held by the top 10 domains. 9,167 of 18,900. Based on 540 prompts across 1,620 runs, measured .
A third of the firms Perplexity consistently names have no page of their own among the sources it cites — the AI knows them from directory listings, not from anything they published.
share of consistently-named firms whose own domain is absent from the cited sources for that slice. 432 of 1,263. Based on 540 prompts across 1,620 runs, measured .
Which domains does Perplexity cite most?
Across 18,900 citations from 540 prompts, 1,749 distinct domains were cited. The distribution is extremely concentrated: the ten domains below account for 48.5% of every citation in the corpus.
| # | Domain | Type | Citations | Share |
|---|---|---|---|---|
| 1 | justia.com | directory | 1,834 | 9.7% |
| 2 | attorneys.superlawyers.com | directory | 1,526 | 8.07% |
| 3 | avvo.com | directory | 1,085 | 5.74% |
| 4 | lawyers.com | directory | 1,082 | 5.72% |
| 5 | expertise.com | directory | 928 | 4.91% |
| 6 | bestlawyers.com | directory | 807 | 4.27% |
| 7 | bestlawfirms.com | directory | 728 | 3.85% |
| 8 | lawyers.findlaw.com | directory | 408 | 2.16% |
| 9 | reddit.com | media | 386 | 2.04% |
| 10 | lawinfo.com | directory | 383 | 2.03% |
| 11 | lawyers.law.cornell.edu | edu | 185 | 0.98% |
| 12 | ontoplist.com | directory | 180 | 0.95% |
| 13 | trustanalytica.org | other | 165 | 0.87% |
| 14 | lawfirmsquare.com | directory | 144 | 0.76% |
| 15 | martindale.com | directory | 139 | 0.74% |
By practice area
Each practice area has its own page with the firms Perplexity Sonar consistently named in each metro, and the sources it cited.
| Practice area | Prompts | Metros | Directory share | First source is a directory |
|---|---|---|---|---|
| criminal defense | 90 | 30 | ≥51.9% | 78.4% |
| employment law | 90 | 30 | ≥53% | 66.5% |
| family law | 90 | 30 | ≥54% | 88.3% |
| immigration | 90 | 30 | ≥50.3% | 72.6% |
| medical malpractice | 90 | 30 | ≥48.1% | 77.4% |
| personal injury | 90 | 30 | ≥54% | 83.7% |
What this report does not show
- Single engine. Perplexity Sonar only; ChatGPT, Gemini and Google AI Overviews are not measured in this period and may differ materially.
- Three runs per prompt. Sonar is non-deterministic; roughly half of all firms named appear in only one of three runs. That is why the published metric is a consistency threshold rather than a percentage share.
- Source classification is incomplete. 36.0% of citation volume sits on domains not yet classified as firm or directory, so the firm-versus-directory split is published as a bound.
- Firm names are extracted from answer prose by pattern matching and alias collapsing, not by a curated firm registry. Name variants may still split.
- Metros, not full state coverage. 30 metros across 23 states; results do not generalise to rural markets.
- No causal claim. Being cited is not the same as being hired, and this report measures visibility only.
Cite this
Based on 540 prompts, measured .
InterCore Technologies, The AI Answer Share Report, 2026-Q3. https://intercore.net/ai-answer-share-report
Further reading
- How this report is measured
- How to get cited by Perplexity
- Generative engine optimization for law firms
- Agentic search optimization
The AI Answer Share Report is published by InterCore Technologies, a marketing agency that works with law firms. We publish it because we needed the numbers ourselves and none existed. Corrections are welcome and will be noted on the methodology page.
