What Indian founders are told when they ask an AI for a cofounder tool

When an engine recommends a product, we ask it to say what for, in two to four short attributes, and we store them. Across the first eight sessions for one account, 628 answers in all, here is what the engines said they were recommending rivals for.

Most of these sessions were on Gemini without search, so they measure the model’s memory more than the live web. The attributes held when search was turned on; the names changed.

Price is the loudest thing in the room

Add the price words together and they outweigh everything else: free tier, free platform, generous free tier, India pricing, INR pricing. An Indian founder asking an AI for a tool is, in the engines’ reading, asking what it costs and whether they can start for nothing. A product with no pricing page is absent from that conversation by construction. The engines cannot recommend you for a free tier they cannot find stated anywhere.

This is the most common finding across the accounts we have looked at, and it is the cheapest fix on the list. One page, first paragraph, the free tier and the rupee price in one sentence, with FAQ markup for “how much does it cost” and “is there a free plan”.

Trust words are about third parties, not about you

Verified profiles, trusted, proven, curated. These attach to products that other pages vouch for. The engines searched for the brand’s name followed by “reviews” four times across the sessions and found nothing, so no trust word could attach. A customers page with named founders and dated quotes helps. So does being listed anywhere at all: the trust words in these answers came mostly from directory and roundup pages, not from the products’ own sites.

“AI cofounder” is not one category to the engines

The names recommended split three ways: cofounder matching sites such as CoFoundersLab and Cofounder Dekho, general tools such as Notion, ChatGPT and Claude, and a small set of virtual-cofounder products such as Venturus AI, FounderPal and zerostartup.in. A brand in the third group that describes itself with the category’s words gets compared with the first two groups, and loses on their attributes. The fix is in the entity sentence on the homepage: say what it is in words that put it in the right group, then say the attributes that group is judged on.

The pages the engines read are the pages to be on

The sources are the other half of the brief. For this category in India the engines read indiacostartup.in, cofounderdekho.in, cofoundercircle.ai, bigideasdb.com, startupindia.gov.in, Medium and Reddit. A listing or a guest piece on each, leading with the attributes above in the engines’ own words, is worth more than any amount of on-site optimisation for the answers that never read your site in the first place.

How this becomes a plan

The attributes group into the page that answers them. Price words need a pricing page. Trust words need a customers page and third-party listings. Guidance words, mentorship and support, need a how-it-works page that shows a person in the loop. India words need to be stated on the homepage and in the Organization markup’s area served. The plan page in Oftheard does this join automatically, against the pages the last site check crawled, so it can say “you have no pricing page” rather than “consider your pricing messaging”.

If every rival wins on free tier and India pricing, that is the page to write. Not a strategy. A page.The plan, reduced to one line