We ran a repeatable set of commercial queries through the major answer engines and recorded which sources they cite. The pattern is not what SEO teams expect — and it is very gameable, in a good way.
- The strategic problem. Search is splitting in two.
- The operating choice. **Finding three: consistency of facts matters more than volume of content.** Where a brand's name, location, founding date, service list and pricing are stated identically across its own site, its business profile and third-party mentions, answer engines describe it confidently and correctly.
- The better model. **What not to do.** Do not mass-produce thin question pages hoping to be scraped: answer engines dislike them and so do readers.
The strategic problem
Search is splitting in two. One half is still ten blue links. The other half is an answer engine that reads the web, forms a view, and cites a handful of sources. For commercial Indian queries — 'best digital marketing agency in India', 'how much does SEO cost in India', 'agency vs in-house marketing team' — that second half is where an increasing share of high-intent research now happens, and almost no Indian brand is optimising for it deliberately.
**What we did.** We ran a fixed set of commercial and research queries through ChatGPT, Perplexity and Google AI Overviews, repeated them across sessions to control for variance, and logged which domains were cited in the answers. This is an observational audit, not a controlled experiment: answer engines are non-deterministic, personalise, and change models without notice. Anyone claiming precise, stable citation share is selling something.
**Finding one: citations skew heavily toward structured, extractable pages.** The sources that get cited disproportionately are pages that state an answer plainly near the top, use clear headings that mirror the question, and contain concrete specifics — numbers, ranges, named steps. Beautifully written narrative pages that bury the answer in paragraph six are cited far less often than plainer pages that answer immediately.
**Finding two: listicles and comparisons dominate commercial queries.** For 'best agency' style questions, answer engines lean on aggregator listicles and directory pages rather than on the agencies' own sites. The agencies that appear in answers are usually those quoted inside third-party lists, not those with the strongest homepage. Getting cited therefore has as much to do with being present, accurately described, and consistently named across third-party sources as with your own on-page SEO.
The operating choice
**Finding three: consistency of facts matters more than volume of content.** Where a brand's name, location, founding date, service list and pricing are stated identically across its own site, its business profile and third-party mentions, answer engines describe it confidently and correctly. Where those facts conflict, engines either hedge or omit the brand entirely. Inconsistent NAP data is now an AI-visibility problem, not just a local-SEO one.
**Finding four: pricing transparency is a citation magnet.** Queries about cost are among the most common commercial questions, and very few Indian agency sites publish real numbers. The handful of pages that state actual ranges get cited repeatedly, because they are the only sources that can answer the question at all. Publishing your pricing is currently one of the cheapest AI-visibility wins available in this category.
**Finding five: recency is weighted, and stale pages decay fast.** Pages with visible, honest dates and recent updates appear more often than older pages covering the same ground. Republishing a genuinely revised page with a new date works; changing the date without changing the content does not, and it damages trust with human readers who notice.
**What to actually do.** Answer the question in the first hundred words of the page. Use headings phrased the way people ask. Publish concrete numbers and ranges you can defend. Make every fact about your business identical everywhere it appears. Maintain a machine-readable summary of your organisation — an llms.txt file and clean structured data both help. Earn accurate mentions in third-party lists, because for commercial queries those are what answer engines read. And update the pages that matter on a real cadence.
The better model
**What not to do.** Do not mass-produce thin question pages hoping to be scraped: answer engines dislike them and so do readers. Do not fabricate statistics to look quotable, because the citation you win is worthless the moment someone checks. And do not treat this as a separate discipline from SEO — the same substance wins both, it just has to be structured so a machine can lift it.
We re-run this audit each quarter. The engines change, but the underlying instruction has been stable so far: be specific, be consistent, be current, and be quotable.
"Answer engines cite the source that can actually answer the question. In Indian marketing, that is a startlingly low bar."
Turn this article into a working growth decision.
Bring us your current numbers, website, and agency reports. We'll show you what is signal, what is theatre, and what deserves budget next.
Book an honest audit →