When someone asks Perplexity "Is [your firm] good?" or "reviews of [your firm]", the engine synthesizes a verdict in seconds from whatever public sources it retrieves: Google reviews, legal directories like Avvo, FindLaw, Lawyers.com and Canadian equivalents such as Law Society directories and Legal Line, news coverage, discipline records, Reddit threads, and your own website. That synthesized paragraph — not your homepage — is increasingly the first impression prospective clients get, and most firms have never read theirs.
The stakes scale with matter value. A consumer choosing among three injury firms may skim the verdict; a general counsel or a family selecting estate counsel will read every cited source. Research on AI-assisted purchasing consistently shows users treating synthesized summaries as a screening step — firms filtered out at that step never learn they were considered, which is what makes an unmanaged AI reputation an invisible leak in the intake funnel.
Reputation management for the AI era means managing the source layer Perplexity reads, because you cannot edit the answer directly. The good news: the inputs are finite and largely controllable. This guide maps how Perplexity builds firm reputations, how to audit yours, and how to shift a mediocre or wrong answer — in both Canadian and US markets. It pairs with our guide on getting recommended by Perplexity, which covers the offensive side of the same system.
Perplexity retrieves live results for a firm-name query, weights sources by authority and freshness, and compresses them into a few sentences with citations. In practice that means:
The synthesis step introduces a distortion worth understanding: compression amplifies extremes. A firm with 95 positive reviews and 5 detailed negative ones may get a verdict that devotes a full sentence to the complaints, because specific, well-written criticism is more quotable than fifty variations of "great lawyer, highly recommend." The defense is not suppressing negatives but out-publishing them with equally specific positive material — detailed reviews that mention the matter type and outcome, case-result pages, and third-party coverage that gives the engine richer positive sentences to draw from.
Note also that reputation answers are regenerated per query, not cached. The verdict a prospect sees today reflects today's retrieval, which is why fixing the source layer changes the answer within crawl cycles rather than requiring any appeal to Perplexity itself — there is no reputation "record" to petition, only sources to improve.
Before auditing anything else, audit the query your prospects actually run. Referral data and intake interviews consistently show that firm-name-plus-"reviews" is the dominant AI reputation query, followed by partner names for higher-value matters. Start there rather than with vanity phrasings your prospects never type.
Run this audit quarterly, in an incognito session: query your firm name alone; firm name + "reviews"; firm name + "complaints"; each named partner; and "[firm name] vs [main competitor]". Log the verdict, every cited source, and any factual errors — wrong practice areas, departed lawyers, outdated addresses, and mis-attributed reviews are all common. Then trace each error to its source: nine times out of ten it is an unclaimed directory profile, a stale Google Business listing, or an old news item outranking newer coverage.
Score the result honestly on three axes: accuracy (is anything false?), sentiment (would a prospect hire you off this paragraph?), and coverage (does the answer reflect your best work, or one loud review?). This mirrors the measurement discipline in our guide to tracking Perplexity citations — reputation queries simply become part of your monthly panel.
For firms with multiple offices or many lawyers, extend the audit matrix: each office location and each client-facing partner generates its own reputation surface, and it is common for a firm's flagship office to have a strong AI verdict while a satellite office inherits a weak one from a sparse Google profile. Prioritize fixes by revenue exposure — the partner who sources the most consultations deserves the cleanest AI answer.
Because Perplexity weights fresh, authoritative sources, you change the answer by changing what it retrieves:
Sequence the fixes by speed of effect. Directory reconciliation and Google Business updates propagate fastest — often within two to four weeks of the next crawl. Review velocity shifts the verdict over one to two quarters. Fresh press and structured self-coverage compound over six to twelve months. A firm with an actively wrong answer (defunct partner listed, wrong practice area) should fix the offending source the same week and then re-run the query fortnightly until the correction appears; a firm with a merely thin answer should invest in the slower, compounding layers first because they raise the ceiling, not just the floor.
Two lines you cannot cross: do not manufacture or incentivize reviews in ways that violate consumer-protection or law society rules, and do not respond to negative reviews with client confidences — regulators in both countries have disciplined lawyers for exactly that. Legitimate removal channels exist for genuinely fake reviews (platform flagging, and defamation remedies in egregious cases), but volume of authentic positive signal is the durable fix.
Reputation defense also has an offensive complement: the more genuinely useful, citable assets your firm publishes, the more of the retrieval surface you own when someone researches you. Practice guides, interactive tools like our legal wizards, and lawyer-authored explainers all show up alongside reviews when Perplexity assembles its verdict, framing your firm as an authority rather than a rating. Firms with a deep content footprint routinely see their own materials cited in reputation answers — which is the strongest position available, since you wrote the source.
Finally, systematize it: add reputation queries to your monthly Perplexity audit, assign review-generation to a defined intake workflow, and recheck directories twice a year. Firms that treat their AI-synthesized reputation as a managed asset consistently outperform those that discover a bad answer only when a lost prospect mentions it. LexScale.ai runs this entire system — audits, review velocity programs, entity cleanup, and structured content — as part of AI SEO engagements for firms across Canada and the US. Book a free strategy call, or start with the broader playbook in the Perplexity for law firms hub.
LexScale.ai builds the content, schema, and entity architecture that gets law firms cited by Perplexity, ChatGPT, Gemini, and Google AI Overviews — across Canada and the United States.
Book a Free Strategy Call →Free Tools & Resources
AI SEO for Law Firms · AI Website Design for Law Firms · Free AI Visibility Grader · AI for Every Practice Area
Ready to grow your firm with AI?