How AI Answers India Beauty Shoppers

Indian shoppers now ask ChatGPT, Perplexity, and Google AI Overviews which sunscreen suits their skin, why their hair is falling, and whether Ayurveda beats a dermatologist's brand. This report reads those answers at category level, side by side with what people in India actually type into Google. It does not rank or score any brand.

What buyers ask AI in this category

Everyday product questions carry the demand

The questions with real search volume behind them are practical and specific: which sunscreen, which anti-dandruff shampoo, which toothpaste, what to do about hair fall or acne. Every one of these product questions traced back to a real, measurable Google search. Searches for the best Indian sunscreens run at about 12,100 a month, and toothpaste searches at a similar 12,100. Men's perfume, at 9,900 a month, sits not far behind.

AI answers to these questions are also the most specific in the category:

• Sunscreen. Every ChatGPT and Google AI Overviews answer to "what matters most to Indian consumers when choosing a sunscreen" raised avoiding a white cast on darker skin tones, tied to light, non-greasy textures that hold up in humid weather. Perplexity mentioned white cast far less often and gave a more general answer about texture and climate.
• Hair fall. The engines agree on the causes (hard water, pollution, diet, hormones) and give shoppers a timeline measured in months, not weeks, before results show.
• Toothpaste. The engines describe it as a habit purchase: people rarely switch, families pass the choice down, and one long-established brand serves as the default example.

Trend questions get confident answers with little search behind them

The picture flips for the industry's own trend vocabulary. "Clean beauty," "vegan beauty," "K-beauty versus Ayurveda," "gender-neutral beauty," and "male grooming trend" all came back with no measurable search demand in India.

The same held for questions about how shopping itself is changing, such as whether people now buy beauty products on Instagram or ask AI instead of Google. About three in four of the questions about trends and changing shopping habits had no measurable demand. That does not mean no one in India cares about clean or vegan products; it means almost no one types those phrases into Google.

The engines answered all of them anyway: fluently, at length, with citations, and in closely matching language ("real but not yet mainstream," "a hybrid, not a replacement"). They rarely noted that few people search this way. Read together, the answers sound like the trade press describing the industry to itself, not like Indian shoppers describing what they want.

Engines are frank about their own blind spots

Asked how well AI chatbots understand Indian skin tones and climate, one Google AI Overviews answer put it plainly: standard AI chatbots and generic skin apps "perform poorly with Indian skin tones and tropical climates because they are mostly trained on Western data." Asked whether AI favors global beauty brands over Indian ones, all three engines pointed to the same cause: global brands have more coverage in the content AI learns from. Google AI Overviews said AI models "rely heavily on massive global training datasets, historical web presence, and widespread digital footprints rather than local market nuance." Perplexity was the most careful, calling AI "not inherently biased toward global brands over Indian ones" while pointing to the same mechanism.

For an Indian brand with strong shelf presence but a thin web footprint, that is the risk in plain terms: AI answers can lag behind the brand's real standing in the market. For a deeper look at whether engines agree with each other, see how AI engines agree on sources before brands.

What a brand in this category should do. Check your positioning words against Google Search data before building campaigns around them. Put your clearest answers on the practical questions shoppers really ask (texture, climate, skin tone, how long results take), because that is where demand and AI answers meet.

Where AI gets its answers

The site types that supply the evidence

The AI answers in this report cited 40,304 sources across 2,443 websites. The evidence is concentrated: 45 sites supplied half of all citations, and the top 10 alone supplied 28.9%. None of those top 10 is a beauty brand's own website. They fall into four kinds of site:

• Creator and community platforms: Instagram, YouTube, and Reddit.
• Retailers and marketplaces: Nykaa, Amazon India, and Smytten.
• Editorial and health publishers: NBC News, Allure, and Healthline.
• Medical bodies: the American Academy of Dermatology (aad.org).

These figures come mostly from ChatGPT and Google AI Overviews answers, which carried the large majority of citations. For what makes a page worth citing, see how to write content that AI engines cite.

Each engine leans on a different mix

Google AI Overviews and ChatGPT answered the same questions but drew on different parts of the web. Google AI Overviews leaned on social platforms and shops, ChatGPT on medical and editorial sources. On Google AI Overviews, Instagram, YouTube, Nykaa, and Amazon India together supplied 28.0% of the sources cited, led by Instagram at 9.5%. On ChatGPT, the American Academy of Dermatology led at 6.0%, followed by NBC News at 5.2%, and its top four sites together supplied 18.7%.

A brand that earns a place in one engine's sources has not earned it in the other's. Strong retailer listings and creator coverage help with Google AI Overviews; clinical evidence and credible editorial coverage matter more for ChatGPT.

No brand owns the answers

The 250 questions built from real Google searches drew answers that mentioned 373 different beauty and personal care brands in late July 2026, counting product brands only, not shops, marketplaces, or clinics. The five most mentioned brands together took 13.2% of those mentions, and 368 other brands shared the rest. No single brand sets the default answer for this category.

The answers keep moving

Brand mentions in these answers shift from one day to the next. Only 7.3% of answers to the same question on the same engine mentioned exactly the same brands the next day, and in 58.8% of those next-day pairs more than half of the brand list changed. When a brand was mentioned, the next answer to the same question on the same engine, usually a day later, mentioned it again 57.9% of the time. Google AI Overviews held its brand lists steadiest of the three engines, and Perplexity changed most.

The reasoning inside the answers moved far less than the brand examples did. What changes from day to day is mostly which brands get used to illustrate an otherwise stable answer. That is why a single check tells you little: one day's mention, or one day's absence, is a weak signal on its own. What one AI visibility check is worth covers this in more detail.

What a brand in this category should do. Map where your category's answers come from on each engine, then invest where the gap is: retailer pages and creator coverage for Google AI Overviews, dermatology-grade evidence and editorial coverage for ChatGPT. Measure presence over weeks, not once.

The market around these answers

India's beauty and personal care market was valued at about $23 billion in FY25 and is projected to reach $40 billion by 2030, according to RedSeer's Beauty Blitz 2030 analysis. Spend per person is still low: about $14 a year in India, against $313 in the US and $38 in China, according to the RedSeer and Peak XV Beauty Unveiled report. AI shopping is already commercial here. Nykaa signed a multi-year partnership with OpenAI in June 2026 to bring its catalogue into ChatGPT as a shopping app.

Survey research points the same way. BCG reported in January 2026 that 64% of Indian consumers use GenAI during their purchase journey. Kantar's Influencer Playbook found in April 2025 that 67% of Indians trust influencer recommendations more than traditional advertising. A global Criteo Health and Beauty Pulse study found 38% of consumers across six markets use AI assistants while shopping for beauty, with no India-only figure.

These surveys measure what shoppers say they do. This report measures what AI engines say, so the two are consistent with each other, and neither proves the other.

What a brand in this category should do. Treat AI answers as a shelf that already exists, not a future channel, and plan category content for it now.

How this report was made

Zumi asked ChatGPT, Perplexity, and Google AI Overviews unbranded shopper questions about Indian beauty and personal care between 23 July and 2 August 2026, and kept the full answer text, the brands each answer mentioned, and the sources it cited, using the same Zumi platform brands use to track their own categories. No question included a brand or asked for a comparison.

Most questions came straight from real Google searches in India. The open-ended shopper questions, written the way people talk to AI, were then checked against the same Google Search data, so AI answers and real search demand could be read side by side as two separate signals. The "3 in 10" finding covers only those open-ended questions.

Limits, stated plainly:

• The findings cover three engines: ChatGPT, Perplexity, and Google AI Overviews. A handful of answers from Gemini, Grok, Copilot, DeepSeek, and Google AI Mode are also counted in the citation totals, adding 40 of the 40,304 citations.
• Google Search data gives rounded monthly averages, not exact counts.
• The window is short, so the day-to-day figures describe late July 2026, not seasonal change.
• This report measures what AI engines say, mention, and cite. It does not measure traffic, sales, or revenue, and nothing in it implies that a mention in an AI answer drives either.

Zumi sells AI search intelligence and has a commercial interest in this category taking AI answers seriously. No customer data appears in this report. Author: Karthick Sreedaran, Co-founder, Zumi.

Zumi tracks what ChatGPT, Perplexity, Google AI Overviews, and other major AI engines say about a category, which sites they cite, and how that changes week to week, so a beauty brand can see where it stands in its own category's answers. Book a demo.