01 Proof, defined

Proof and case notes — what we can show, and what we won't claim

Most case-study pages in this category are fiction with logos. This page is built the other way around: the standard of evidence comes first, and the stories are only added once they meet it.

Answer

ProminAI publishes what it can evidence and states plainly what it cannot. Every deliverable is logged in the client portal with a dated live link. Observations of AI platform responses are recorded verbatim with the date checked. ProminAI does not publish visibility scores, guaranteed rankings, or client outcomes it cannot independently verify.

  • Dated, clickable evidence
  • No invented case studies
  • No visibility scores

02 Three levels

What “proof” means at ProminAI

Most marketing reports blur three different kinds of statement into one number. We keep them separate, name them differently, and hold each to a different standard — because a claim you cannot classify is a claim you cannot check.

  1. 1

    Proof of work

    Fully guaranteed

    The article exists at this URL, on this date. The mention exists at this URL, on this date. Clickable, verifiable, contractual.

  2. 2

    Observation

    Recorded, not guaranteed

    On this date, this platform returned this text for this question. True as a record — but platform behaviour varies by user, session, region and time.

  3. 3

    Outcome

    Never guaranteed

    Leads, sales, rankings, citations. Influenced by many factors ProminAI does not control — and therefore never promised, by us or honestly by anyone.

Each level of proof mapped to what ProminAI promises, how you verify it and what could change it
Level What we promise How you verify it What could change it
1 · Proof of work Every plan deliverable, delivered and documented with a dated live link — or made good at no cost. Click it from your portal, any day of your plan. Screenshot it. Download the list. If a host ever takes a page down, the deliverable stops counting until we replace it — the count only ever reflects what is live.
2 · Observation Verbatim, dated records of what each selected platform returned for your tracked questions — baseline first, then on schedule. Ask the same question on the same platform yourself. It takes a minute and costs nothing. Platforms refresh their models and sources on their own cycles; the same question can return different answers to different people.
3 · Outcome Nothing. We will not promise outcomes, and we put that refusal in writing. Your own enquiries, bookings and sales records — which stay yours, not ours to claim. Market conditions, seasonality, pricing, competitors, your own follow-up — the whole business, not one programme.

Every number on this site, in your reports and in your portal belongs to exactly one of these levels — and is labelled with it. When a vendor quotes you a “visibility score”, ask them which level it belongs to. There is no honest answer.

03 Case notes

Case notes — none yet, and here is why that is the honest state

We’re publishing these as programmes complete. ProminAI launched recently. Rather than dress up early work as case studies, we’re waiting until we have full-cycle observations we can show with dates and links. In the meantime, the ₹3,499 Starter exists precisely so you can judge the quality of the work yourself, at low risk, before committing to a programme.

Try it first — ₹3,499

What every case note here will contain

The template is fixed before the first story is told — so the format can never bend to flatter a result. When a note appears on this page, this is its exact shape:

  1. Anonymised descriptor

    Business type, region, plan and length — e.g. “A multi-speciality dental clinic, Delhi NCR — Momentum, 3 months”. Names only with written permission.

  2. Starting position

    Which failure mode the baseline audit found, and what public evidence was missing on day one.

  3. What was built

    Assets and mentions, with counts — the same counts the client’s portal logged, nothing re-totalled for marketing.

  4. What was observed

    Verbatim platform observations with the date of each check. Quotes, not summaries.

  5. What changed — and what did not

    Both, always. A case note that only reports movement is an advertisement.

  6. Limitations

    On every single note: “These are observations on specific dates. AI platform responses vary and are not controlled by ProminAI.”

One more rule: client names and logos appear only with written permission, and anonymised notes stay anonymised even when the story is flattering. If a case note ever names a client, the permission exists in writing — that is the only way a name gets here.

04 Realistic timelines

How long before AI visibility actually changes?

Answer

You will see published assets and logged mentions within the first month — that part is immediate and verifiable. Changes in how AI platforms describe a brand depend on how thin the starting position is, how competitive the category is, and how often each platform refreshes. Three months is the shortest period in which meaningful change can usually be observed. Six to twelve months is where category association genuinely builds.

Month by month: what is built, what can be observed, and what still cannot be claimed
Period What is built What can be observed What still cannot be claimed
Month 1 Brand Truth File approved; first article live; first verified mentions logged; baseline audit records what every selected platform says today. Every deliverable, clickable in your portal — and the honest starting picture, however thin. Any change in AI answers. Platform refresh cycles have not passed; anyone claiming movement this early is describing luck or fiction.
Month 3 A body of published articles and a growing, verified mention base; profiles corrected and consistent. First differences against the baseline on some tracked questions, on some platforms — recorded verbatim, with dates. Consistent presence across platforms, or any business outcome. Early movement is a signal, not a settlement.
Month 6 Coverage across your priority questions; corroboration deep enough that platforms cross-confirm the same facts. Clearer category association on more questions; competitor movement tracked alongside yours, both directions. Permanence, or inclusion in every answer. Platforms keep refreshing; the work keeps compounding — neither stops.
Month 12 A full-year public evidence base: the durable asset the programme exists to create. Where category and consistency allow it, the brand described accurately and repeatedly across refresh cycles. A guaranteed position — ever. That claim is not available honestly, from anyone, at any budget.

Anyone who tells you two weeks is guessing.

The two questions this table always raises

What if nothing changes?

Then you will see that clearly in your reports — and you will still hold everything you paid for: the published articles, the verified mentions, the corrected profiles, the Brand Truth File. These are permanent public assets, not rented attention, and they keep working after the programme ends.

Honestly: outcomes vary by category, competition and starting point. What does not vary is the deliverable guarantee — if anything in your plan is not delivered and documented with a live link, we make it good at no cost. And your reports include competitor movement, so you always see the full picture, not a curated one.

Can you guarantee my brand appears in ChatGPT?

No — and no one can. ProminAI will never promise a ranking or position inside ChatGPT, Gemini, Perplexity, Claude or Google, inclusion in every answer, a permanent citation, or a guaranteed number of leads. Nobody can guarantee the behaviour of a platform they do not own. Anyone who offers to is selling something that does not exist.

What we do guarantee: every deliverable in your plan, delivered and documented with a live link — or we make it good at no cost. We guarantee the work and the proof. We will never guarantee a ranking, because no one can.

05 The wider evidence

What the wider evidence shows

These are independent, published findings about how AI search is changing buying behaviour — the reason this category exists. Read them for what they are: none of these figures is a ProminAI client result, and we will never present industry data as if it were.

≈4.4×better conversion from AI-referred visitors than ordinary organic searchSOURCE: SEMRUSH
94%of business buyers used a gen-AI tool in their most recent purchase processSOURCE: 6SENSE
60%of Indian online searchers begin at least one search a day with an AI assistantSOURCE: YOUGOV · APRIL–MAY 2026
73%of Indians use AI to help with shopping and buying decisionsSOURCE: ADOBE
86%of Indian consumers likely to use gen-AI for at least half of purchase decisionsSOURCE: ACCENTURE CONSUMER PULSE 2026, INDIA · 2026
≈3×more likely to be cited by AI systems — brands with 4+ matched sameAs profiles vs 0–1SOURCE: PUBLISHED ENTITY-SEO RESEARCH · 2026
~23%more AI citations for sites with a well-structured llms.txt than matched controlsSOURCE: PRINCETON GEO-BENCH · MAY 2026

Two of these findings shape the work itself: entity corroboration is why the programme builds matched, verified profiles rather than one-off posts, and the llms.txt finding is why machine-readable brand files are part of the technical layer — including on this very site.

Every figure, its source, its date and its link — on the data & sources page

06 Judge us on work

The cheapest way to judge us is to make us prove it once

The ₹3,499 Starter is one complete, published authority article: researched, written to the same standard as every membership article, approved by you, and delivered with its live link. No programme commitment, and the full ₹3,499 is adjusted against your membership if you upgrade within 30 days. You are not buying a promise — you are buying one round of evidence and judging it.

Try it first — ₹3,499 Credited in full against any membership within 30 days.

Next step

Start with what AI says about you today

The free AI Presence Check records what selected platforms return for your real buying questions — dated, in writing. It is the same baseline discipline every case note on this page will be measured against.

Not ready for a conversation? Start with the ₹3,499 Starter and judge the work first.