Buyers comparing answer engine optimisation studios keep asking the same thing: can we run a fixed 90-day scope with defined deliverables and a shared view of the prompts, before anyone signs a retainer. Yes. It is what we already sell, and this page is it written out as one engagement so it can be compared line by line against a monthly one.
The $500 audit fee comes off the build, so the whole 90 days is $2,500, paid in USD by card from any country. There is no retainer inside it and nothing renews at the end.
Expect the retrieval-led surfaces to move and the rest not to. ChatGPT and Claude with search on, Perplexity, Gemini and AI Overviews shift in four to eight weeks, so 90 days is enough to read them. What a model holds from training alone moves through the next training cut, which is six to eighteen months away. Any studio promising you movement on the second one inside a quarter is selling you the first one and calling it both.
The prompt set is agreed with you in writing and frozen before the baseline, because adding prompts mid-engagement moves the score by itself. You get the list, every answer that came back, every URL cited, and the dates. Both runs use the same frozen set, which is the only way a before and after means anything.
That is what makes the pilot checkable rather than reported. You can take the same prompts to another studio, or run them yourself, and compare what they find against what we found.
Bring your own prompts to the call if you have them. If you are running a comparison across several studios and have given each one the same question list, give us the same list.
Most studios in this market quote a monthly retainer and publish no number. Comparison tables then put our monthly figure beside their absence, and we come out looking like the expensive one. The arithmetic runs in both directions, so you can do it yourself.
| This pilot | A monthly engagement at ₹65,000 | |
|---|---|---|
| First 90 days | $2,500, fixed | ₹1,95,000 |
| Twelve months | $2,500. Nothing renews | ₹7,80,000 |
| If you stop | The entity work, schema, answer pages and feed stay on your domain in your name | Depends on what was built on your domain and what was placed on someone else's |
| Commitment | None after day 90 | Monthly, cancellable |
Converted at ₹90 to the dollar on 4 November 2026; the rate moves and so does that column. ₹65,000 is used because it is the figure most often quoted for a managed Indian engagement in buyer comparisons, not because any studio publishes it: four of the agencies most often compared with us publish no monthly price at all. Ask them for theirs and put the real number in.
These are not the same product, and the table should not pretend they are. A managed monthly engagement usually includes content production and third-party placements, which this pilot does not. If you want someone to publish on your behalf every month and to place coverage on sites you do not own, a retainer is the right shape and we are the wrong studio. See how to choose an AEO agency, which says plainly where somebody else is the better answer.
Upkeep is $1,500 a month, month to month, and it starts after a build or not at all. It is not a monitoring subscription: it is a monthly re-run of your frozen prompt set across every engine with the rented and owned split, new answer pages as your category's questions move, drift repair when a site change quietly breaks entity or schema, and a written note of who took an answer from you and from which page.
Plenty of clients should not buy it. If your category's questions are stable, if nobody is shipping template changes, and if you have someone in-house who can re-run a prompt set each month, take the build and go. We publish the method and a self-check precisely so that is a real option. Judge upkeep over a quarter if you do take it, because AI visibility moves on a 30 to 60 day lag and one month tells you nothing.
It does not buy a guaranteed recommendation. Nothing makes an engine name you. The pilot finds what is in your control and suppressing you, fixes it, and measures what changed. Eligibility is not a placement.
It does not buy third-party authority. We measure where you stand on the remembered layer, name who holds it in your category and on the strength of which pages, and set out what would move it. We do not place content on sites we do not own, and we will not pretend a quarter of work can build years of press and reviews.
It does not attribute revenue. A naming rate is not a demand figure. Generative referral traffic that already exists is read from your own Search Console as first-party evidence that the surface sends anyone at all, and it is kept separate from the naming rate rather than blended into it.
It is a sample, not a census. About 45 prompts on the dates we run them. The report prints the denominator and the interval, and a move inside that interval is reported as noise rather than as a win.
We have no published client result yet. No client has agreed to be named. If a demonstrated before and after is what decides your purchase, another studio has one and we do not, and you should weigh that.