Answer Engine Optimisation is how D2C beauty brands show up inside ChatGPT, Claude, Perplexity and Google's AI Overviews when a customer asks 'best niacinamide serum for Indian oily skin', 'which vitamin C serum is dermatologist-approved under ₹1,500' or 'how do I layer retinol and hyaluronic acid'. The answer is no longer ten blue links - it is a synthesised paragraph that names three to seven brands, and the brands that get named are the ones with the right entity data, enough review velocity across enough platforms, and content shaped for LLM extraction rather than for classic SERP ranking. Baclinc has run the AEO playbook on its own domain - a documented llms.txt, structured FAQ coverage, entity-hardened Organization schema, and an active digital PR pipeline - and we've adapted the same discipline for client work. The honest caveat repeats - no public D2C beauty AEO case study yet, because the practice itself is 18 months old and the measurable share-of-voice inside LLM answers for any Indian category is a rolling moving target. What we bring is a first-principles approach - LLMs cite brands with clean, referenceable entity data and high review velocity - not a mystical 'AI SEO' service that sells vibes.
AI-driven search in the beauty category is moving faster than most agencies are tooled for. Google AI Overviews, which rolled out broadly in 2024, now return for a large share of skincare queries in India - 'how to treat pigmentation', 'best sunscreen for oily skin', 'retinol for beginners' - and the overview typically names 3-6 brands. ChatGPT and Claude responses to 'best [ingredient] for [concern] in India' queries surface a different but overlapping set, and Perplexity's citation-first format surfaces a narrower, more source-weighted set. Gemini's behaviour is inconsistent but rising. Across all four, three patterns are consistent. First, the brands cited have high entity consistency - same founders, same HQ, same founding year, same product list across Google, Wikidata, Crunchbase, LinkedIn and their own Organization schema. Second, the brands cited have measurable review velocity - not just review count. A brand with 2,000 lifetime Google reviews but zero new reviews in 60 days is cited less than a brand with 400 reviews and 15 new ones last month. Third, the brands cited are referenced on authority third-party surfaces - Vogue listicles, Cosmopolitan roundups, MakeupAndBeauty reviews, dermatologist blogs - because LLMs weight second-party citations heavily in a YMYL category where they themselves hedge against hallucination. The practical implication: AEO work is 30% technical (schema, llms.txt, entity alignment), 30% content (FAQ depth, extractable Q&A shapes), and 40% off-site (review velocity, listicle placement, authoritative brand data distribution). Agencies selling pure 'technical AEO' are selling a third of the work. The industry is also 18 months old as a named practice, so we're honest about what's measurable and what's still early - our tracking stack gives a real share-of-voice view but the baseline numbers in Indian beauty are volatile week to week, and the honest report reflects that.
