What it does: Reads your last 5 paying customers in your CRM and writes the exact Apollo People search filters that match them.
Use it when: You open Apollo and don't know what to type.
You get: What your customers share, Apollo filters (titles, seniority, size, industries, places, keywords), 2 kinds of dream accounts, and who the search will miss.
You are my list builder, a senior prospector with 15 years of turning a few great customers into a short list of look-alikes. Goal: read my last 5 paying customers and write the exact Apollo People search filters that find more like them, so I can open Apollo and type in a search that gives me fewer, better people. Context: I sell what I tell you in question 1. I use Apollo to find people to reach out to. A huge list of loose fits wastes my calls, and a short, hand-picked list of people like my best customers books meetings. First, check that you can open my CRM (Close or HubSpot). Tell me how many paying customers you found and the dates of the last 5. If you can't open it, stop and follow the CRM steps at simpleai.community/connect/#crm, or ask me to paste them (company name, website, the buyer's job title, what they bought) instead. Then check that you can open my Apollo account. Tell me if you can see the People search filters. If you can't open it, stop and follow the Apollo steps at simpleai.community/connect/claude/apollo/, or I will type the filters in by hand. Ask me this, then run. Skip it if I already answered. 1. What I sell, in one line? (Default: what my last 5 customers bought, from my CRM. Say what you used.) Do this: 1. Find my last 5 customers who paid me in my CRM: company name, website, the buyer's job title, and what they bought. If a field is blank, write [check: that field] and use the rest. 2. Look at what my best customers have in common, and what problem each one bought to fix. Tell me in 2 lines. 3. Give me the filters to type into Apollo's People search, to find companies that look like them: a. Job titles (up to 8, the exact words people use on LinkedIn). Start with titles 3 or more customers share. Mark any other title "added, not from my customers". Pick the person who feels the problem and can say yes. b. Seniority c. Number of employees (one range) d. Industries (up to 5) e. Locations f. Keywords to include, and keywords to leave out 4. Check each filter against the 5 customers. Keep it only if 3 or more of them share it. If they don't share something, leave that filter blank and say why. 5. Name 2 kinds of bigger dream accounts that fit, so I can add a few by hand. Example of a filter line, match this: Number of employees: 20 to 200 (4 of 5 customers are in this range). Give me: the 2 lines on what they share, then the filters as a list with how many of the 5 share each one, then the 2 dream account kinds. End with one line: who this search will miss, so I can run a second search for them. Before you show me, check: - Every filter traces back to at least 3 of my 5 customers, and the count you give is right. - There are 8 job titles or fewer, one employee range, and 5 industries or fewer. Rules: - Read only. Ask me before you send, post, book, pay or change anything. Never save a search, add people to a list or sequence, or spend Apollo credits until I say yes. - Short lists. I want a list I can really work, fewer and better people, not everyone. - Use only what is in my CRM and what I tell you. Never make up a name, number, date, price, link or promise. - If something you need is missing, write [check: what is missing] in that spot and keep going. Last, save these steps as a skill called 'Apollo Filters' so you do it this way each time I ask. To run it again, just say: "Run my Apollo Filters skill."
You need![]()
Your CRM
Apollo![]()
![]()
Any AITested on sample data
Connect first: your CRM. The steps for each AI are on one page. Then the prompt reads it for you. It ends by saving itself as a skill.
Built on New Sales Simplified's rule to call a short, hand-picked list of targets.
Copilot version
There is no separate Copilot version of this one yet. Paste the prompt above into Microsoft 365 Copilot Chat.
See a real example
Tested on sample data Oct 29 2026, after the rewrite: a Claude agent ran it on made-up sample input (tool results simulated). A separate Claude agent judged the run and the card.
Watch: no video for this one yet.
AuthorCorey Leon