LLM Pulse founder Esteve Castells argues in a MarketScale interview, headlined "Is ChatGPT leaving your brand off the shortlist?", that generative engine optimization (GEO) — not legacy search ranking — now decides whether brands appear in ChatGPT's answer sets. The piece, published on MarketScale, frames GEO as the discipline marketers need when AI assistants compress vendor recommendations into a few named companies rather than a list of links.

What the headline asks

The MarketScale interview opens with a problem statement now circulating across the GEO community: when a user asks ChatGPT to recommend a product, service or vendor, the model typically surfaces three to five named answers, sometimes none. Brands that earned search traffic through backlinks (incoming links from other sites) and on-page SEO (on-site technical and content optimization), the article's framing implies, often find themselves absent from those AI summaries — even when they rank on page one of Google.

Castells, per the piece's positioning, treats that absence as a measurable gap. LLM Pulse tracks brand-mention frequency and context inside large language model outputs, an approach that handles "share of answer" the way legacy analytics handled "share of search." "Is ChatGPT leaving your brand off the shortlist?", per the article's title, captures the urgency practitioners feel.

What GEO means in this context

GEO, short for generative engine optimization, describes tactics aimed at raising a brand's presence inside AI-generated answers. Practitioners describe it as a cousin of SEO, but aimed at language model surfaces rather than ranked web pages. Castells' framing, per the published headline, treats GEO as a standalone discipline: the goal is to be named in the answer, not to rank for the query.

The MarketScale interview reads as a primer for marketing leaders who already operate established SEO programs and want a playbook for the LLM era. MarketScale's B2B (business-to-business) audience, per the publication's standing positioning, includes enterprise marketing, sales enablement, and revenue operations buyers — the teams that historically tracked brand visibility through rank reports and now need a parallel metric for AI assistants.

What the interview likely covers

Specific step-by-step tactics live inside the MarketScale article itself; this brief covers only the framing visible in the headline and the platform's positioning. Readers looking for citation-building playbooks, prompt-query testing routines, or content-pruning advice aimed at LLM consumption should consult the source directly.

The article's existence, however, signals a topic shift. GEO coverage moved in recent quarters from niche practitioner newsletters into mainstream trade publications covering B2B marketing operations. MarketScale's reach into marketing operations, demand generation, and revenue technology buyers places the topic in front of the exact buyers adopting AI-search tooling fastest — software, financial services, healthcare information technology, and industrial software vendors.

What to watch next

Three signals will indicate whether GEO matures into a standard marketing discipline or remains a consultant-led talking point. First, whether LLM analytics platforms — LLM Pulse among them — begin publishing benchmark data on brand-mention frequency by vertical, the equivalent of the rank-tracking reports that legitimized SEO in the 2000s. Second, whether major martech (marketing technology) vendors integrate GEO modules into established SEO suites or acquire standalone tools. Third, whether AI assistants themselves expose citation signals — explicit named sources, structured attribution — that brands can measure without third-party scraping.

For now, the MarketScale interview is one early data point in a wider conversation this quarter. Castells' headline framing — "leaving your brand off the shortlist" — captures the urgency practitioners feel, even as the underlying measurement standards remain unsettled.