Sign inJoin the waitlist

Indian sources won the Indian questions: upgrowth, themediaverse, nicodigital and hawkorbit across 189 readings

The India cut of our 15 October 2026 benchmark: 189 readings of ChatGPT and Gemini, four Indian agency domains cited 52, 48, 44 and 33 times, and 2 citations of our own domain.

Published by AI Knows Us · Updated 15 October 2026

On the Indian questions in our 15 October 2026 benchmark, ChatGPT and Gemini cited Indian websites. Across 189 India readings, upgrowth.in was cited in 52, themediaverse.in in 48, nicodigital.com in 44 and hawkorbit.com in 33, which puts small Indian agencies above several global specialist platforms on the same questions. Our own domain was cited twice in those 189 readings, both times on ChatGPT. The practical finding for an Indian business is simple: on an Indian question these two engines go looking for an Indian source, and there are very few of them publishing anything measured.

The full three market run sits in what AI assistants actually cite across 597 readings of ChatGPT and Gemini in three markets. This page is the India slice and what an Indian publisher can do with it.

The India readings, counted

Date run15 October 2026
EnginesChatGPT and Gemini
Questions102, of which 72 are worded identically across the three markets and 30 are India specific
India readings189: 102 on ChatGPT, 87 usable on Gemini
Readings citing aiknowsus.com2 on ChatGPT, 0 on Gemini
Market controlfixed in the prompt text; the India prompt came back answered for Mumbai

Note the 87. The run attempted 612 readings and used 597, and all 15 of the readings we could not use sit in this one cell, Gemini in India. Every India figure above is read against 189 rather than 204, which is why the denominator is printed next to it.

Which Indian websites the engines used

DomainReadings citing it
upgrowth.in52
themediaverse.in48
nicodigital.com44
hawkorbit.com33

Set those against the specialist platforms the same run recorded: Peec at 49 readings, Scrunch at 11 and Rankscale at 3. upgrowth.in at 52 was cited in more readings than any of the three. hawkorbit.com at 33, which is a small Indian agency, was cited in more readings than Scrunch and Rankscale put together. None of these counts is a judgement on anybody's product. They are counts of whose page an engine used while answering a buying question.

India has the most locally specific winners of the three markets we measured. The United States cut produced no US only domain of this kind at all.

Why that is an opening rather than a curiosity

Across the whole run, 85.6% of the 10,652 citations recorded went to a page on the vendor's or the agency's own website. So the Indian domains above were not selected by a directory or a reviewer. They published their own pages and the engines used them. In a market where the number of Indian sites publishing dated, specific, measured work on any given commercial subject is small, that is reachable in weeks rather than years.

What an Indian page needs on it to be the one cited

  • Prices in rupees, with a date. 102 of the 102 build spec conversations we ran asked for pricing in the reader's own currency. A monthly figure in rupees, with the date it was true and what is included, answers the question an Indian buyer is actually asking. Our own pricing page prices by location, so a reader in India sees rupee amounts.
  • A stated sample size. 80 of 102 asked for the N. Ten clients is a number. A page that says it works for everybody is not.
  • A disclosed method. 99 of 102 asked how the thing was measured, which costs a paragraph and makes every figure on the page usable.
  • The engines named. 98 of 102 asked for the engines to be enumerated rather than called AI. ChatGPT and Gemini, on 15 October 2026, beats any sentence beginning with the words all AI platforms.
  • Indian specifics rather than a global page with a city dropped in. GST on the invoice, the city, the language the buyer searches in, what an Indian supplier actually charges. The tier by tier figures in what each SEO tier costs per month in India are an example of the level of detail that gets used.

The India questions that are already settled

Anything asking how the engines themselves work belongs to the engines. developers.google.com was cited in 178 readings across the run and help.openai.com in 112, and the India readings on those subjects went the same way. Review directories took 0.8% of all citations, which is the measured answer to whether an Indian software company should spend a year on directory profiles before publishing anything of its own.

Our own India result belongs in the same list of plain facts. Two readings out of 189 cited us, on ChatGPT. Three of the surfaces our product reads, Google AI Overviews, Claude and Perplexity, are absent from this dataset, so these counts describe ChatGPT and Gemini only.

The India index, and what is not published yet

These 189 readings are a cut of a three market benchmark, not a full picture of Indian AI search. A larger India specific study, broken down by category, by city and by language, is in progress and the page for it carries no numbers yet: the India AI Citation Index says so in those words. Until it is published, the figures on this page and on the benchmark are what we are prepared to stand behind for India.

Questions about the India cut

Why 189 readings in India when the other markets have 204?

Gemini returned 87 usable readings of 102 in India. Those 15 are the whole difference between the 612 readings attempted in this run and the 597 used.

Were the India questions asked in Hindi or other Indian languages?

This run was in English, with the market locked to India in the prompt text, and the answers came back for Mumbai. Language level breakdowns are part of the India index rather than this dataset.

How many India readings cited your own domain?

Two of 189, both on ChatGPT. Gemini did not cite us in any market in this run.

Do Indian agencies get cited because of their location?

They get cited on Indian questions. upgrowth.in, themediaverse.in, nicodigital.com and hawkorbit.com were cited in 52, 48, 44 and 33 readings, and 85.6% of citations across the run went to self published pages, so what the engines used was their own content about the Indian version of the question.

When is the India figure taken again?

After 60 days, on the same frozen 102 questions, with a fresh date and N attached to the result.

What to publish first if you sell in India

Pick the one question your Indian customers ask before they buy, answer it with rupee figures, a date, a sample size and your method, and publish it on your own domain rather than waiting for somebody to list you. That is what the four Indian domains above had in common, and the count next to each of them is public now so anyone can check how low the bar is.

Being straight about our own position: AI Knows Us sells AI visibility measurement, and we were cited in 2 of 189 India readings in our own run. The figures on this page come out of that run, the method is published, and the India index page will say the same when it has numbers on it.

Read next

A 30 Day AI Visibility Plan for an Indian Small Business, and What Two Readings Can Honestly Tell YouA dated four week sequence built against the Indian field as it was measured in October 2026, where the sites holding the answers are small Indian domains and one engine column starts at zero.AI visibility for Indian manufacturers and industrial suppliers: the certificate number and its validity date are the citable fact85.6% of the 10,652 citations we recorded across 597 ChatGPT and Gemini readings on 15 October 2026 went to somebody's own website, and the quality file in your office is the page nobody has typed up.AI visibility for Indian professional services firms: the named specialism is the entity, not the firm85.6% of the 10,652 citations we recorded across 597 ChatGPT and Gemini readings on 15 October 2026 went to somebody's own website, and a full service page gives an assistant nothing to file you under.AI visibility for Indian ecommerce brands: the declarations you already print are the product dataMRP, net quantity, country of origin and the returns window are the facts a shopper asks an assistant for, and Indian law already makes you publish them. Most stores publish them as a photograph.AI Visibility Audit Deliverables for a UK Agency: The Five Artefacts, and Why the Zero Column Is One of ThemWhat a British agency should hand a client at the end of an audit, including the findings that did not survive a second reading.A Workflow Standard for AI Search Optimisation: Six Stations, and Who Owns Each OneThe execution sequence an Indian agency can hold a team to, measured against what our own site serves eleven named crawlers.

Or browse the whole archive.

See what AI says about you.

The first scan is free, it needs no account, and it asks the questions your buyers ask.

Free. No card. We ask 5 real buyer questions on 2 AI apps.

Already sure you want this? Join the waitlist. We open on 31 October 2026. It costs an email address. No card, and nothing is charged. Or talk to a person if you have a question first. Or read how we fix what it finds.