AI prompt template
AI prompt to find buying signals on a company site
A signal is an event on a public page with a date and a reason it matters to you. The prompt returns those and nothing that is only inferred.
When to use it
- You want to rank a list by timing, not just fit.
- You have the careers, news and pricing pages for each company.
- You will use the result in the first line.
The prompt
Paste it into the model you use, then replace each bracketed input.
We sell: [WHAT_WE_SELL] Look for these kinds of events: [SIGNAL_TYPES] Using only the pages below, list each event you find. For each: what happened, the date if the page states one (otherwise "date not found"), a verbatim quote, the page address, and one sentence on why it matters to our offer, written as a possibility, never as a fact. Order them by how recent they are. Never invent an event or a date. If you find no events of these kinds, return: not found. [PAGES]
| Input | What to put there |
|---|---|
[WHAT_WE_SELL] | One sentence on your offer. |
[SIGNAL_TYPES] | The events that would make you relevant: hiring for a role, a new office, a pricing change. |
[PAGES] | Page text with addresses. |
Why it works
- Requiring a date and a quote separates a signal from a hunch.
- The “why it matters” is marked as a possibility so it cannot harden into a claim in the email.
- An empty result is a valid, useful answer.
An example
Illustrative. The companies and pages below are invented.
Input
WHAT_WE_SELL: rostering for hotels SIGNAL_TYPES: a new property, hiring for a general manager PAGES: tamarindhospitality.example (illustrative)
Output
Event: a second hotel. Date: date not found. Quote: "Tamarind Hospitality now operates two hotels in the city." Page: tamarindhospitality.example/about. Why it might matter: two sites can make shared rostering relevant. Event: general manager role open. Date: not found. Quote: "General Manager, Tamarind Riverside" Page: tamarindhospitality.example/careers. (Illustrative output on an invented company.)
Variants
Only the last ninety days
Using only the pages below, list events dated within the last ninety days (today is [TODAY]) that are relevant to [WHAT_WE_SELL]. Each with the date, a verbatim quote and page address. Never invent a date; if a date is not stated, skip the event. If none, return "not found". [PAGES]
Hiring only
List the open roles on the pages below that relate to [FUNCTION]. For each: the exact title, the quote, the page address, and the manager it reports to if the posting says so, otherwise "not found". Never infer. [PAGES]
Limits
What are the limits of the find buying signals on a company site prompt?
Related
Research an account from its own pages
A prompt that turns a prospect’s own pages into sourced notes for a cold email: what they do, what changed, who owns the problem, each with a quote.
Write a sourced first line
A prompt that writes a first line for a cold email from one quoted fact on the prospect’s page, refusing to flatter or guess when nothing is there.
Pick the right buyer from public titles
A prompt that picks who to write to from the people and roles a company’s own pages name, and says “not found” when no page supports a choice.
Cold email templates
Templates by situation and by role.
Research and writing, with a quote behind every fact
Outlevo runs these steps for each company it finds, and shows the page each fact came from.