The Birth Injury SEO & AI Search Playbook
Two distribution systems now sit between a query and a firm: the results page that lists, and the answer engine that summarises. They reward overlapping but different things. This is how to build for both, measure them apart, and avoid the specific failure of writing pages that rank and are never cited.
Five parts, roughly a twenty-five-minute read. Published in full on this page. There is no gated version, no download wall, and no form. If you want it as a PDF, print the page — it is styled for it.
Part 1 — Two systems, one page
A results page ranks documents. An answer engine extracts and synthesises passages, then cites the sources it drew from. Both start from the same page, and both reward quality — but the unit of value differs, and that changes how a page is written.
| Classic results page | Answer engine | |
|---|---|---|
| Unit of value | The document | The passage |
| Wins by | Being the best result for the query | Being the most liftable accurate statement |
| Rewards | Depth, links, internal structure, freshness, page experience | Self-containment, real tables, accurate schema, unambiguous claims |
| Punishes | Thin content, duplication, poor experience | Ambiguity, claims that need surrounding context, inaccurate structured data |
| Measured by | Impressions, clicks, position | Citations and referral sessions by engine |
The practical synthesis: write the answer first, then the argument. A page that opens with two paragraphs of framing before it says anything definite ranks acceptably and gets quoted rarely.
Part 2 — The extraction checklist
| Requirement | Test | Why |
|---|---|---|
| A direct definitional answer in the first paragraph | Can the first 40 words stand alone as an answer to the target query? | That paragraph is what gets extracted |
| Self-contained factual blocks | Lift any paragraph out. Does it still mean the same thing? | Synthesis destroys context; self-containment survives it |
| Real tables with headers | View source: is it <table> and <th>, or divs? | Tabular data is only parsed as tabular data if it is marked up as such |
| Accurate schema, not stuffed | Does every schema claim match a visible on-page fact? | Inaccurate schema teaches an engine to distrust the whole domain |
| Named author and real review date | Is there a person, with credentials, and a date that reflects an actual review? | Accountability signals carry weight in health and legal topics |
| An llms.txt | Does it state what the site claims and what it does not? | A machine-readable statement of scope and standing caveats |
| Primary sources linked inline | Does every statistic link to the dataset or statute it came from? | A claim an engine can verify is a claim it can safely repeat |
This site is held to that checklist and its llms.txt is public, including a standing caveat about the CMS obstetric data blackout — which is exactly the kind of statement an answer engine should be able to pick up and repeat.
Part 3 — What to publish, and in what order
A birth injury firm’s organic programme has two audiences that must not be mixed on one domain: firm-side commercial intent, and family-side informational intent. Mixing them creates a conflict that is visible to a regulator and confusing to both readers.
| Tier | Content | Intent | Order |
|---|---|---|---|
| 1 | Practice-area cornerstone pages: what the firm does, how a matter proceeds, what a records review involves | Informational, family-facing | First — they earn the links |
| 2 | Jurisdictional pages, where the firm can say something true and specific about that state | Informational and local | Second, and only where there is something distinct to say |
| 3 | Answer-shaped pages targeting the actual questions families ask | Informational | Continuous |
| 4 | Attorney and firm authority pages with real credentials | Navigational and trust | Early, then maintained |
| 5 | Referral and co-counsel pages aimed at other lawyers | Commercial, firm-facing | Underbuilt by almost everyone; high value |
Do not build fifty templated jurisdiction pages. A find-and-replace page set is both a search-quality risk and a credibility one — the first partner who reads two of them notices. Publish a jurisdiction page only where the underlying data or law lets you say something that is true of that place and not of the next one. We hold ourselves to the same rule; see how many state pages we actually publish and why.
Part 4 — Measurement
| Channel | Leading indicator | Lagging indicator | Setup |
|---|---|---|---|
| Classic organic | Non-branded impressions and clicks on target clusters | Signed cases attributed to organic | Search Console plus analytics, cluster-tagged |
| Answer engines | Citations and referral sessions by named engine | Signed cases with an assistant touchpoint | Named referral sources for each assistant, configured before launch |
| Brand | Branded query volume | Direct inquiries | A lagging measure of everything else you do |
Rankings belong in an appendix. A ranking is a leading indicator of a leading indicator, and reporting it at the top is how organic programmes survive for eighteen months without producing a case.
Part 5 — The editorial standard that makes it work
In a category touching both health and legal outcomes, content quality is not a style question. The practical standard is short and hard to fake.
- A named author with real credentials on every substantive page. Not "the firm", not "admin", not nobody.
- A named reviewer — clinical for anything clinical, legal for anything legal-adjacent — with credentials and a review date on the page.
- A last-reviewed date that is real. Automatically bumping dates on unchanged content is a trust liability dressed as a tactic.
- Primary sources only. Statute, dataset, peer-reviewed study. Never another marketing agency’s blog — that is how you inherit somebody else’s error and then defend it.
- Publish to a standard, not a calendar. If the standard yields two pages this month, publish two.
Our own version of this standard, and the build script that enforces it, are published at Editorial Standards.
Sources
- Google Search Central — Creating helpful, reliable, people-first content — includes guidance on experience, expertise, authoritativeness and trust
- Google Search Central — Structured data general guidelines
- Schema.org — vocabulary reference
Cite this analysis
Free to reproduce with attribution, including commercially. Charts may be embedded as published.
Birth Injury Marketing. “The Birth Injury SEO & AI Search Playbook.” 24 August 2026. https://birthinjurymarketing.com/library/birth-injury-seo-and-ai-search/
Journalists and researchers: the underlying data is already published as CSV at /data/, free and without conditions — no form. Need a different cut, by state, county, hospital or year? Ask, and we will also tell you what the data cannot support.
Other publications
The Birth Injury Case Acquisition Playbook
The whole system, front to back: sizing the market, buying demand, screening it, and measuring to the signed case.
The Medical Screening & Intake Manual
The operational system around the constraint that actually limits case volume. The document nobody else has written.
Legal Advertising Compliance for Birth Injury Marketing
A framework for staying inside bar advertising rules in the most heavily scrutinised category in legal marketing.
All eight → · The tools that come from these → · Have us run it →
Authorship & review
- Author
- Kevin Schwaner, Founder; Chief Marketing Officer. Credentials.
- Last reviewed
- Legal review
- Not yet assigned. This page describes marketing practice, not law, and gives no legal advice. Rules cited are linked to the primary source so you can read them directly. See Editorial Standards.
- Corrections
- Found an error? Tell us and it goes in the dated corrections log, whether or not it flatters us.