An AI-visibility proposal you can put your name on.
The structure of a proposal for the client you want to sell this to: what the situation is, what gets measured, what you will do, how you report, what it costs and what you are not promising. Every number below is invented and marked as an example — replace them with your own.
The situation
One short paragraph in the client’s own words: why this came up now. Use something they said to you, not a market statistic you cannot source. If you only have a suspicion, say it is a suspicion — the audit is what turns it into something readable.
Fernpost sells project software to small design studios. In the last two quarters three inbound leads mentioned that they had asked an AI assistant for options before booking a demo. Nobody at Fernpost knows what those answers say, whether Fernpost is named in them, or which sources the assistants are reading to produce them.
What we measure
Two parts. The scope is yours to set — how many questions, which assistants, which competitors. The definition underneath it is not: keep it as written, so the client reads the same description of the measurement that the reports will use.
What we report instead is visibility: the share of sampled answers that name the brand, per assistant. Assistants are shown side by side, never blended into one score.
No figure from fewer than three answers per question per assistant. The range around a share is a 95% interval worked out from the number of answers behind it. Few answers, wide range. Each assistant is asked through its developer API with web search switched on, not through the consumer app. What a person sees in the app can differ.
Microsoft Copilot and Google AI Overviews / AI Mode offer no public API, so they are not measured, and no report estimates them.
| Scope | Proposed |
|---|---|
| Buyer questions tracked | 24 |
| Assistants asked | ChatGPT, Perplexity, Grok |
| Answers per question per assistant, per run | 3 |
| How often | every two weeks |
| Competitors compared | 4 |
| First report | after the second run |
What we will do
Before the first audit you do not know which pieces of work matter, so list the kinds of work and say plainly that the order comes from the measurement. After the audit, replace this with the ranked list the diagnosis produced, each item with its reason. Effort is your estimate for your own team — write it as one.
| Work | Effort |
|---|---|
| Get the client listed and described correctly on the review sites assistants cite | ~12h |
| Write the comparison pages the cited sources keep sending buyers to | ~16h |
| Fix the product pages that answer the questions buyers actually ask | ~10h |
| Monthly reading of the measurements and the next sprint | ~4h / month |
This audit is a single measurement of how assistants answer today. The ranked work is what the current sources suggest, with expected effort — not a forecast of results.
The product reads what assistants already answer and tells you what it found. Anything that changes a client's site stays a human decision.
How we report
Say what arrives, how often, and what the client is asked to do with it. The approval step is worth naming: it is the moment the client agrees to the next sprint, and it is the reason the report is a decision document rather than a dashboard nobody opens.
- A report each month, as a link in our brand, plus a PDF if you prefer to forward a document.
- It shows where you stand now against the start of the period, per assistant, each figure with its range and how many answers sit behind it.
- It lists what we did, what followed, and what we propose to do next.
- You approve the next sprint on the report page by typing your name. Nothing starts before that.
When a piece of work is marked done, the product compares the topics it touched with the client's other topics that it left alone, over the same weeks, and shows the difference as an estimate with a confidence level. It is a record of what followed, not a claim about why.
Models typically take weeks to re-crawl and shift citations, so a short period shows early signal rather than settled results.
What it costs
Your price, your terms. Two things are worth stating next to the number: what the retainer buys each month, and the shortest period after which the measurements say anything at all — quoting a one-month term for something that moves over a quarter sets the client up to be disappointed.
| Item | Amount |
|---|---|
| Set-up: questions, competitors, first audit | $2,400 once |
| Monthly retainer: measurement, diagnosis, the work above, reporting | $3,200 |
| Minimum term before the first honest read | 3 months |
Retainer and effort are the agency's own estimates for the scope below, shown so the numbers behind the proposal are visible.
Models re-crawl and re-cite over weeks. Sixty to ninety days is the honest unit here; a short period shows early signal rather than settled results.
What we do not promise
Keep this section. It costs you nothing that was ever deliverable, and it is the part that makes the rest of the document believable — particularly to a client who has been sold an AI-search package before.
- A rank or position in any assistant
- A single 0–100 visibility score
- That a piece of work produced a change: we show what followed, with a comparison where one exists
- A forecast of the share a client will reach
- A revenue figure for visibility
- Anything about assistants we do not measure
With one client and no untouched topics to compare against, movement cannot be separated from platform-wide drift, and the report says so where that is the case.
Visibility is a share of answers. What that share is worth belongs to your client's model, not to ours.
The next step
One step, with a name against it and a date. If you have run the free audit already, the next step is a conversation about what it found; if you have not, the next step is running it.
- [Your name] runs the audit on Fernpost and the four competitors — no cost, nothing published.
- We walk through what the assistants answered and which sources they cited.
- If it is worth doing, we start on [date] with the scope in section 3.
Template. Example agency, client, competitors and figures are invented.
Fill it in with a real measurement
Section 3 is guesswork until an audit has run. The audit costs nothing, produces the ranked work with a reason on each item, and gives you the report to attach to this proposal.