Answer Engine Optimization for Spanish Law Firms in 2026: What It Is, Why It Matters, How We Measure It

For over a decade, getting found online meant ranking on Google. The website that won page one of a search result also won the click. That world is ending. When a foreign property buyer in 2026 wants to understand Spanish wealth tax, they no longer type “spanish wealth tax expats” into Google and read the first five results. They ask ChatGPT, or Claude, or Google’s own AI Overview, and they read the single synthesised answer the assistant returns. The firm whose content was used to compose that answer wins the relationship. The firms whose content was not used disappear.

What AEO Is and How It Differs from SEO

Search Engine Optimization (SEO) is the work of making a webpage rank highly in a list of links. Answer Engine Optimization (AEO) is the work of making a webpage’s content be extracted, cited, or paraphrased inside an AI-generated answer. They overlap, but the differences matter.

SEO rewards keyword targeting and link authority. AEO rewards structured factual density, schema markup, reviewer attribution, and presence on third-party sources that LLMs actually consult during training and retrieval. An SEO-strong page can be AEO-invisible if its key facts are wrapped in marketing prose; an AEO-strong page can be SEO-modest if its citable data is dense and its schema is correct.

HubSpot launched its AEO product on 14 April 2026 at $50 per month and measures two metrics most agencies are not yet tracking: Presence Quality (does the firm appear in third-party sources LLMs consult — LinkedIn, directories, Reddit, press) and Sentiment Score (how do LLMs actually characterise the firm when asked). These metrics are the entire game in 2026 because they predict whether an assistant will recommend you when a client asks, not whether your blog post ranks when a client searches.

The Ten Signals We Score

At CostaLuz Lawyers we run our own AEO/GEO scorer in-house. It is not a HubSpot subscription wrapper — it is a Python module that probes a page, a domain and an LLM directly and produces a deterministic score out of 140. Eight signals are on-page; two are off-page. We use it on our own site and on every prospective client we audit.

On-page signals (max 100)

• FAQ schema present (20 pts). Pages that expose FAQPage JSON-LD with clean Question/Answer pairs are disproportionately extracted into AI answers.
• Article schema with recent dateModified (15 pts). LLMs that use retrieval favour content explicitly dated within the last 90 days.
• Person reviewedBy (15 pts). A named human reviewer with a linked profile is the strongest E-E-A-T signal an article can ship.
• Direct-answer lead (10 pts). The first 100 words contain a literal Q/A pair or an H2 question followed by a concrete-fact paragraph.
• Citable data density (10 pts). Currency amounts, percentages, dated law references, named statutes per 1,000 words above a fixed threshold.
• Organization schema (10 pts). The LegalService or Organization block sits in the head and matches the rendered author.
• HowTo schema (10 pts). Procedural content uses HowTo/Step JSON-LD where applicable.
• Internal hub links (10 pts). At least two links to canonical resource hub pages on the same domain.

Off-page signals (max 40)

• Presence Quality (20 pts). Brand SERP presence (5), legal-directory mentions (5), LinkedIn firm profile with ≥50 followers (5), and ≥3 unique external domains citing the firm in their content (5).
• LLM Sentiment (20 pts). A real probe: we ask Claude what it knows about the firm, parse the answer, and award +10 for a positive mention with correct specialty, +10 if the practice areas it associates with the firm match reality, and −10 if a competitor is recommended in place of the firm. The result is cached per domain for 30 days; we do not page-spam the assistant.

What This Means for Your Firm in Practice

Most Spanish law firms today score between 35 and 70 out of 140. The bottlenecks are predictable. Schema markup is missing or wrong. Reviewer attribution is generic (“our legal team”) instead of a named person with a linked bio. The first 100 words of every article are throat-clearing marketing prose instead of a usable answer. LinkedIn is abandoned. Third-party directory profiles are stale.

None of this is mysterious. All of it is fixable in eight to twelve weeks of focused work. The firms that fix it first capture the AI-answer market while the rest are still arguing about whether AEO is real.

How CostaLuz Lawyers Can Help

CostaLuz Lawyers is a law firm first and an AEO practitioner second. We do AEO work for ourselves because foreign property buyers find us through AI assistants today, and we have measured the cost of being missing from those answers. The tooling we built for our own site — the 140-point scorer, the deterministic schema injection, the third-party presence map — is the same tooling we now offer to other Spanish firms that want to stop being invisible inside AI answers.

Methodology: this article describes the v2 (2026-05-27) AEO/GEO scorer implemented in CostaLuz Lawyers’ in-house SEO system. The two off-page signals (Presence Quality, LLM Sentiment) were added in response to HubSpot’s launch of its AEO product on 14 April 2026. The scoring weights, signal definitions and the cache TTL are documented in our public standards file docs/standards/seo-geo-system-completeness-v3.md.