The context
Smart TVs are a high-consideration, high-ticket purchase, and the research happens before anyone walks into a store. Buyers ask an AI assistant which panel to get under a budget, which one handles gaming, which brand’s software holds up after two years. The answer they get is assembled from published editorial, not from the brand’s own marketing.
This brand had the market share. It did not have the answer.
The challenge
Present in the category, absent from the verdict.
The opening Beacon audit ran the brand’s high-intent smart TV queries across four AI engines, spanning price bands, features, use cases and brand-trust questions.
The brand appeared regularly in answers and was almost never the recommendation. Competitors held the territory that matters once a buyer has decided to spend — picture quality, gaming, content ecosystem — and the brand’s own website was barely present in the citations behind category answers.
The gap was not awareness. The engines knew the brand. They had nothing citable to justify choosing it.
The strategy
Map the queries before writing anything
Beacon fixed the baseline: which queries the brand won, which it lost, and which it was absent from entirely. Content was commissioned against the losses, not against a content calendar.
Concentrate on a small set of narratives
Rather than chase every query, the brand picked a handful of themes to own outright and put sustained volume behind each one. Engines settle on a shortlist of three to five brands per prompt, so winning a few narratives decisively beats appearing thinly across all of them.
Commission editorial in the formats engines lift from
Long-form reviews, buying guides, head-to-head comparisons and category explainers across Tier-1 and Tier-2 Indian tech media — built with clear headings, specification tables and a stated verdict rather than a hedge.
Re-audit monthly and reallocate
Beacon re-ran the same query set every month. Publisher spend moved toward the formats and narratives that were actually shifting the answers.
The execution
Over six months the programme placed editorial across a mix of Tier-1 and Tier-2 Indian tech titles, weighted toward the query themes where the brand had been losing.
Coverage was deliberately split across tiers. Tier-1 titles carry the authority weight; Tier-2 titles supply the volume and the long-tail query matches. Engines cite both, and repetition across independent sources is what moves a mention into a recommendation.
The outcome
By the closing audit the brand’s share of voice had risen 5 percentage points and its recommendation rate by 6 percentage points, measured on the same query set and the same four engines as the baseline. The clearer shift was in which queries it won: the brand moved from price-anchored visibility into feature and use-case narratives where it had previously not appeared at all.
Measured on the same query set, the same four engines and the same prompt phrasing as the opening audit, re-run monthly.
What we learned
Share of voice and recommendation rate are separate problems and they move at different speeds. In our audit work across consumer electronics we routinely find brands with plenty of mentions and almost no endorsements. Volume of coverage lifts the first number. Format and verdict clarity lift the second.
The other lesson is about patience. Engines re-weight their sources only after the source mix has shifted and stayed shifted. A short burst of publisher content changes very little on its own; sustained placement across independent titles is what changes the answer.
- AI Visibility Audit (Beacon)
- Publisher network strategy
- AEO content commissioning
- Monthly AI visibility reporting