Known by name. Invisible by category.

The buyer doesn't start by typing "Zakro."

They start by typing what's wrong.

Their GST billing and their warehouse register are out of sync. They lose two hours a day to manual invoicing. Stock runs out with no warning.

Those are the questions a real prospect types into ChatGPT or Perplexity before they have ever heard of Zakro.

So this baseline tests two different things at once. Does Zakro show up when someone already knows to ask for it by name? And does it show up when someone doesn't know it exists yet, and is just describing the problem?

Those turned out to be very different answers.

Cited 2 times out of 10 — and the 2 are the wrong 2 to be proud of.

Zakro is cited the moment a prospect already knows its name. Its brand query resolves cleanly, and a head-to-head against the market's incumbent comes out in Zakro's favor for factory-first buyers.

Zakro is invisible on all eight questions that describe the problem instead of naming the product.

Nobody types a brand-name query for a company they've never heard of.

The two prompts Zakro wins are only reachable by someone who already found Zakro somewhere else — a referral, an ad, a sales call. They prove the product is describable and competitive. They do not generate demand.

The eight prompts Zakro loses are the ones an AI engine hands to a prospect discovering solutions for the first time. That is where most of the market is at Zakro's stage.

Across those eight, three kinds of competitor keep appearing: the incumbent everyone already has an opinion about, the general-purpose cloud accounting tools, and a long tail of vertical-specific tools that surface exactly where their vertical is named.

That last group is the tell. Most of them are smaller than Zakro. They get cited anyway — because a source about them exists somewhere the engines already trust for that specific question. That is a content-and-placement gap, not a product gap.

The citable proof already existed. It was just never published where a model could find it.

Zakro's own case-studies page is live and indexed. It states, in plain text, that no case studies have been published yet.

Meanwhile the homepage carries the exact material a model would quote: two hours a day saved on invoicing, ₹1.8 lakh recovered in a single week, zero stockouts after switching.

The content exists. It was never shaped into a page an engine's retrieval can find and lift. That is the first thing the engagement fixes.

We publish the baseline, then re-run the same ten prompts.

This is a Day-0 reading, not a finished story. The number above is the starting line.

The same ten prompts get re-tested after the first round of citable content goes live, so movement is measured against a real baseline instead of claimed. When that re-test runs, this page gets a second chapter — whichever direction the number moves.