Ask ChatGPT for the best vitamin C serum in India, or the best earbuds under a certain budget. It hands you a shortlist. The common assumption is that it built that list by reading Amazon. It did not.
We classified every source ChatGPT cited across the Indian D2C consumer categories in the August 2026 Cited Index, 742 citations across skincare, audio, grooming, baby care, health, pet care, and travel. The picture upends the marketplace assumption:
| Source type | Share of ChatGPT's D2C citations |
|---|---|
| Editorial (tech and lifestyle publishers) | 28% |
| Brand websites | 27% |
| Community (Reddit, YouTube, Instagram) | 23% |
| Marketplaces (Amazon, Nykaa) | 8% |
| Review platforms | 7% |
| Other | 6% |
Editorial is the single largest source. Community is nearly as large as the brand's own site. And marketplaces, the surface most D2C teams treat as the center of the universe, are 8%. When ChatGPT recommends a consumer product, it is mostly repeating what publishers and communities said about it, not what a marketplace listing claims.
The mirror image of B2B
This is worth holding next to how ChatGPT treats software. For B2B SaaS, 72% of ChatGPT's citations are brand sites: you win by publishing. For D2C, the brand's own site is 27% and most of the answer is assembled from places you cannot edit. It is the same engine reading a different web, and what you sell decides which one you get. We mapped that full split in whether AI reads your website depends on what you sell.
For a consumer brand, that means getting recommended by ChatGPT is mostly an off-site job. You cannot write your way onto a publisher's roundup or a Reddit thread from your own domain. You have to earn in.
The editorial 28%: get into the roundups
The largest single source ChatGPT reads for D2C is editorial, the "best [category] in India 2026" articles that tech and lifestyle publishers write. When your brand is in those roundups, ChatGPT reads the roundup and repeats you. When it is not, you are invisible to more than a quarter of the citations before you start.
Most consumer brands have a PR motion built for launches and funding announcements, and no motion at all for getting included in category comparison and recommendation content. On ChatGPT's actual diet, the second is worth far more. Target the publishers your category's buyers read, and pitch to be included in the evergreen "best of" pieces, not just the news.
The community 23%: Reddit, YouTube, Instagram
Want to know how your brand scores on these same metrics?
We'll run 20 prompts across 3 AI platforms and send your report within 24 hours.
Almost a quarter of ChatGPT's D2C citations are community and social. The specific mix shifts by category, and it is worth knowing yours: skincare leans Instagram and Reddit, audio leans YouTube and Reddit, and the pattern differs again for baby care and grooming. We break the per-category source mix down in the skincare AI-visibility playbook, the first of the category guides.
The work here is not a campaign. It is genuine presence: honest answers in the subreddits your buyers use, product seeding with the review creators they watch, and enough Instagram and YouTube coverage that when a model looks for what people say about your category, your name is in the conversation. Hard-selling backfires, because the community turns on it and the model reads the negative sentiment.
The 8% you are over-weighting
Marketplaces are 8% of what ChatGPT cites for D2C. They are not nothing, and they still matter enormously for conversion once a buyer has decided to look. But as a lever for getting into the ChatGPT recommendation, a marketplace listing is a minor input. A brand that has beautiful Amazon A+ content and no editorial or community footprint is optimizing the smallest of the five sources.
The playbook, in priority order
- Earn editorial inclusion (28%). Get into the category roundups real publishers write. This is the largest source and the one most brands ignore.
- Build genuine community presence (23%). Reddit, YouTube, and Instagram, weighted to your category's actual mix. Presence, not promotion.
- Keep your own pages clean and extractable (27%). Your site is a real quarter of the citations. Make product facts, ingredients, and use cases easy for a model to lift. Just do not expect the site alone to carry it.
- Treat marketplaces as conversion, not citation (8%). Keep the listings strong for buyers; do not mistake them for your ChatGPT strategy.
One honest caveat
One limit to hold onto. These shares describe the whole D2C market, every brand pooled, not what will happen to your brand specifically, and they describe where ChatGPT looks, not proof that a single roundup or Reddit thread flips a given answer. ChatGPT is also just one of five engines, and Gemini in particular reads your own site far more, so do not port this plan straight across. Confirm it on your own category: run your buyers' prompts through each engine and read the citations.
But the correction is the useful part: getting your D2C brand recommended by ChatGPT is an editorial and community job first, a website job second, and a marketplace job a distant fourth. Most brands have that order exactly backwards.
Methodology. 742 ChatGPT citations across seven Indian D2C categories (Skincare & Beauty, Audio & Wearables, Personal Grooming, Baby Care, Health & Wellness, Pet Care, Travel & Luggage) in the August 2026 Cited Index, classified by source-domain type and pooled by citation count. Aggregate and market-wide: no individual brand is named, and this is a different measurement system from the per-brand Citation Mix in the Cited dashboard. Citation Source Mix is the sixth of the Cited 8 metrics.
Want to see where AI cites your brand? Get a free AI Visibility Report across ChatGPT, Gemini, and Perplexity.