Why Corporate Law Firms Lose Search for Their Own Name

When a client's in-house counsel asks ChatGPT about a corporate law firm's mergers and acquisitions practice, the firm competing for that answer is not the other law firm down the street. It is Chambers. It is Legal 500. It is Hukumonline's Practice Leaders directory. A documented 2026 case study queried "Cravath, Swaine & Moore M&A expertise insights," a search that should have surfaced the firm's own thought leadership first. Of the six sources the AI cited, four belonged to directories the firm does not control, and only two were Cravath's own pages.

That finding, reported in the 5W/Haute Lawyer Network 2026 Legal AI Visibility Report, is not a quirk of one prestigious American firm. It is the structural default for corporate legal search, and Indonesian firms sit inside the same structure. Hukumonline's Practice Leaders directory profiles hundreds of firms across dozens of practice areas. Chambers and Legal 500 rank the same corporate mandates internationally. A firm that invests heavily in regulatory briefings and thought leadership but never touches its own entity infrastructure is publishing content into a system that has already decided the directory speaks for it.

Why Ranking and Being Cited Are Not the Same Job

Search engine optimisation for most industries assumes that better content, more backlinks and cleaner technical implementation move a page up a results list. That model still applies to corporate legal search, but it answers only half the question. The other half is whether an AI system treats the page as a verifiable enough source to cite when it answers a question about the firm by name.

Those are different mechanisms. A directory listing does not need to write better prose than a law firm's own website. It needs structured, consistent, third-party-verified data: firm name, practice areas, partner credentials, matter categories, all in a machine-readable format an AI system can cross-check. A law firm's own site usually has the substantive content and almost none of the verification infrastructure. The directory wins the citation not because its writing is better, but because it is easier to trust automatically.

The Scale That Makes This Worth Fixing

The addressable market behind this problem is large enough that a small citation-share gain compounds meaningfully. Southeast Asia's legal services market was valued at roughly USD 30.6 billion in 2024, with IMARC projecting growth to USD 47.9 billion by 2033. Mordor Intelligence separately sizes the broader Asia-Pacific legal services market at approximately USD 128.71 billion for 2025, within a global legal services market Polaris Market Research places at roughly USD 1,032 billion the same year. Indonesia's own contribution sits inside that regional figure without a single authoritative country-level number, since most detailed Indonesia-specific legal market sizing remains behind paywalled industry reports, but Indonesia's macro conditions, 5.11% GDP growth in 2025 per BPS data and 74.6% internet penetration among roughly 212 million users per DataReportal, both point toward a corporate legal market with substantial room to grow into.

Two further data points sharpen why the ranking-versus-citation distinction matters specifically now. A 2025 GrowthSRC study tracking roughly 200,000 keywords found position-one organic click-through rate fell from 28% to 19% year over year on queries where an AI Overview appears, a direct erosion of the value a top ranking alone delivers. First Page Sage's July 2026 update still finds the top three organic results capturing approximately 68.7% of all clicks, meaning ranking well remains necessary. It has simply stopped being sufficient on its own, which is the entire argument for treating entity and citation work as a parallel discipline rather than an optional add-on to conventional SEO.

Structured data compounds this effect measurably. Sites implementing FAQPage schema correctly have been documented achieving citation rates roughly 2.7 times higher than equivalent pages without it, a gap large enough that skipping schema implementation on an otherwise strong page represents one of the more avoidable losses in the entire citation-eligibility picture described below.

The supply side of this market adds one more reason entity clarity matters specifically in Indonesia. PERADI reported not less than 70,000 registered advocates as of its 20th anniversary in December 2024, spread across 53 regional advocate organisations operating under a single national bar mandate. That fragmentation, more organisational bodies than most single-bar jurisdictions maintain, makes consistent entity data across a firm's own site, its directory profiles and its bar registration more valuable as a differentiator than in a market where one central body issues and displays every credential in one place.

What Makes a Page Citable, Not Just Well Written

The signals that determine whether an AI system will cite a law firm's own page are the same signals Google's quality raters use to evaluate Your Money or Your Life content, applied one level more literally. Anonymous content cannot be verified, so it is not cited. A named advocate with a linked, credentialed bio can be. A claim without a statutory citation cannot be checked against a primary source, so it carries less weight than one that quotes UU No. 18/2003 Pasal 22 directly.

Named, credentialed authorship is the starting point. Every substantive page should carry the name of the reviewing advocate, linked to a bio page that lists PERADI registration, education and practice focus. The sameAs field in that bio's Person schema, pointing to the bar directory entry, creates an externally verifiable credential chain, the legal-sector equivalent of a physician's licence number appearing on medical content.

Primary source citation density follows close behind. Content that cites legislation, court decisions and regulatory guidance by document number and article is substantially more citable than content that paraphrases the same point without attribution. This is not a stylistic preference. Aggarwal et al.'s KDD 2024 study, the Princeton GEO research widely cited across the field, found that content built with authoritative citations and clear entity signals earned roughly 40% more citations in generative engine responses than unoptimised content covering the same topic.

The Schema Priority Order Corporate Firms Usually Get Backwards

Most corporate law firm websites, when they implement structured data at all, start with the firm-level LegalService schema and stop. That is the wrong order of priority for a firm competing against directories that already have entity-level structured data at scale. Attorney and partner schema, connecting each named lawyer to verifiable credentials, is the piece that actually differentiates a firm's own site from a directory's aggregated listing.

Notice what is absent from that list: the deprecated Attorney type that Schema.org has since replaced with the more inclusive LegalService and Person combination. Firms that implemented schema several years ago and have not revisited it are frequently still running the deprecated version, which is functionally invisible to the newer entity models AI systems increasingly rely on.

Directory Presence Is Additive, Not a Competitor to Defeat

The instinct to treat Hukumonline, Chambers and Legal 500 as competitors to outrank is understandable and largely counterproductive. A listing in Hukumonline's Practice Leaders directory functions as a contextual citation from the highest-authority domain in the Indonesian legal vertical, the rough equivalent of a hospital appearing on Alodokter in the healthcare space. Firms that maintain a well-optimised independent site alongside complete directory profiles capture both the aggregator's authority halo and the direct-traffic preference of clients who evaluate multiple sources before making contact.

Firms that neglect their own site and rely solely on directory listings lose something specific: the E-E-A-T credentialing opportunity that attorney-level schema and owned educational content provide. A directory profile can confirm that a firm exists and ranks. It cannot demonstrate, in the depth an AI system rewards, that a named partner has handled a specific class of cross-border transaction, cited the applicable regulation correctly, and kept that analysis current as the law changed.

The Advertising Rule That Shapes Every Lever Above

None of this entity work operates outside Indonesia's advocate advertising restrictions. Kode Etik Advokat Indonesia (KEAI) Pasal 8(b) prohibits advertising designed solely to attract attention, and Pasal 8(f) prohibits seeking publicity through mass media about specific matters. Pasal 4(c) prohibits promising an outcome. Practically, this rules out comparative superlatives ("the leading M&A practice in Jakarta") and case-outcome claims of any kind, on a webpage or inside a schema field.

What it does not rule out, and in fact actively encourages, is exactly the content described above: named credentialed authorship, statutory citation, factual practice descriptions, and educational material. UU Advokat Pasal 22 treats legal education as part of an advocate's professional duty. The compliant path and the AI-citable path converge on the same content, which means the entity and schema work is not a workaround for the advertising restrictions. It is the direct, permitted expression of them.

What This Looks Like Applied to a Real Practice

Consider a firm with a genuinely strong cross-border M&A practice, three partners with real transaction experience, and a website that describes all of it in well-written prose with no schema, no linked credentials, and no statutory citation. That firm will rank reasonably well for its own branded searches and remain almost entirely absent from AI-generated answers about its own practice area, because nothing on the page gives a machine a way to verify what the prose claims.

The fix is not a rewrite. It is entity infrastructure layered onto content that is often already good: Person schema for each partner linked to their PERADI registration, LegalService schema connecting the firm to its practice areas, statutory citations added to existing regulatory briefings, and a directory audit confirming that Hukumonline, Chambers and Legal 500 profiles say the same thing the website does. None of it requires new marketing claims KEAI would restrict. All of it is the difference between being the two citations out of six and being the four.

---

Frequently Asked Questions

Does implementing schema markup violate KEAI's advertising restrictions?

Not when the schema describes visible, factual page content: credentials, practice areas, contact details. The open question, not yet resolved by PERADI, any Indonesian court, or the Dewan Kehormatan, is whether schema specifically designed to trigger a "best" or "top" recommendation crosses the line. Firms should implement schema for informational and educational content and avoid configurations aimed at competitive superlative queries.

How long does it take for schema and entity work to affect AI citation?

Technical implementation shows up in Google's Rich Results Test and structured data reports within days. Whether an AI system starts citing the firm depends on the platform's own crawl and training cycle, which varies from roughly two hours for Perplexity to several weeks for ChatGPT, based on documented platform behaviour differences.

Should a firm try to outrank Hukumonline or Chambers directly?

No. Both function as citation infrastructure a firm's profile needs to be complete on, not competitors to defeat in search results. The realistic objective is appearing alongside them in an AI-generated answer, not replacing them.

Is the deprecated Attorney schema type actually a problem, or just outdated documentation?

It is a functional problem. Schema.org's deprecation signals that newer entity-recognition models are built around the LegalService and Person combination. A site still running Attorney-type markup is not automatically broken, but it is not benefiting from the schema improvements the newer types were introduced to support.

What is the fastest single change a corporate firm can make?

Auditing every partner bio page for a linked PERADI registration number and Person schema. It is the single most effective E-E-A-T fix identified across the research, and it typically requires no new content, only structured data added to pages that already exist.

For the deeper mechanics of how AI systems weigh directory citations against a firm's own entity signals, see our companion piece on Generative Engine Optimization for legal services, and for the full compliance framework behind every claim above, our Legal SEO practice builds every engagement around it. A fuller treatment of E-E-A-T, entity clarity and citable content architecture runs through Cited or Silent: The Definitive GEO, AEO & AI Visibility Playbook, free to read with registration.

---