Trestle

You are my data checker.

Goal: keep only real mobile numbers on my call list. Dead numbers burn my best calling hours.
My Trestle results: [paste or upload the export]

Do this:
1. Keep only numbers that came back as mobile.
2. Count how many were landline, fixed internet and other internet numbers.
3. Keep the rest of each row the same.
4. Tell me what share of the list was mobile, and how many dials that leaves me.

Rules:
- Keep only line type Mobile. Ignore the activity score. It is not a cut.
- Never move an office number into the mobile column, and never call an office number a mobile.
- Count only what is in the results. Never round or guess a number.
- Use only what is in my data 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.

Give me the counts first, then the clean list with the same columns.

Last, save these steps as a skill called 'Keep Real Mobiles' so you do it this way each time I ask.

Built from: Fanatical Prospecting, Jeb Blount, Chapter 6: Know Your Numbers

You are my list cleaner, a data pro who preps call lists for phone teams.

Goal: one list split into CALL, CHECK and CUT, so I can dial only numbers that pick up.
Context: my business dials lists every day. Bad numbers waste hours, and the people on the list want calls only on real lines.

First, check that you can read the Trestle results file I uploaded. Trestle has no connector, so export it as a CSV first. Tell me how many rows you found. If there's no file, stop and ask me to paste the CSV instead.

Sort every row:
- CALL: a mobile with an activity score over 30.
- CHECK: a landline or internet phone line with a score over 30. These may be a front desk. Call them last.
- CUT: a score of 30 or less, or a number Trestle says is not valid.

Give me:
1. Counts for each group, and the share of the list each one is.
2. The CALL list, sorted by score, high to low: name, company, phone, line type, score.
3. The CHECK list the same way.
4. The CUT list with the reason for each row.

Give me each list as a table. Keep the reasons under 8 words each.

Rules:
- The score checks the line, not the person. Say this once at the top.
- Line type decides the group first. A mobile over 30 goes in CALL. A landline or internet line over 30 goes in CHECK, even if its score is high. The score only sets the order inside a group.
- If one person has 2 numbers, sort each number by the rules above. Then mark the one in CHECK as "second number". If both are in the same group, mark the lower score.
- If a row has no score, put it in CHECK and say "no score".
- Every row goes in exactly one group. The 3 counts must add up to the rows I pasted. Check that before you answer.
- Use only what is in my data 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, say so, and keep going.
- Do not send, post, or change anything without my yes.

Before you show me, check:
- Check that CALL plus CHECK plus CUT equals the rows in the file.
- Make sure every landline or internet line is in CHECK or CUT, never in CALL.

Last, save these steps as a skill called 'Clean My Phone List' so you do it this way each time I clean a list before dialing.
To run it again, just say: "Run my Clean My Phone List skill."

Built on Fanatical Prospecting's rule to know your numbers.

You are my numbers checker, a dialer analyst who has studied pickup rates for years. You are a pro at this.

Goal: a table of pickup rate by Trestle score band, so I know which numbers are worth my time.
Context: my business pays for phone checks. I want to know if the score is worth it on my own list.

First, check that you can read my Trestle file and open my call log in my CRM or dialer. Tell me how many rows each has. If you can't open the call log, stop and tell me to connect it (Claude: Customize, then Connectors. ChatGPT: Settings, then Plugins), or ask me to paste both files instead.

Trestle has no connector, so export its results as a file and upload it.

File 1, my Trestle results: phone, line type, activity score.
File 2, my call log for the same numbers: phone, number of dials, picked up yes or no, talked over 60 seconds yes or no.

If I don't say, use the last 90 days of calls.

Do this:
1. Match the files by phone number. Say how many matched and how many didn't.
2. Put the matched numbers in score bands: 0 to 30, 31 to 50, 51 to 70, 71 to 90, 91 to 100.
3. For each band: numbers, dials, pickups, pickup rate (pickups divided by dials), talks over 60 seconds.
4. Do the same split for mobile, landline and internet line.
5. Tell me the lowest band still worth dialing on my list, and how many dials I'd have saved by cutting below it. If a band has under 30 numbers, still give the band's numbers, but mark the cut-off as [check: too few numbers in this band to judge] and say what more data would settle it.

Rules:
- Show your math in the table so I can check it.
- This is my list and my time. Don't call it a rule for every list.
- Use only what is in my data 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, say so, and keep going.
- Don't send, post or change anything without my yes.

Before you show me, check:
- Check that matched plus unmatched equals the numbers in each file.
- Make sure every pickup rate in the table is pickups divided by dials.

Last, save these steps as a skill called 'Check My Pickup Scores' so you do it this way each time I compare phone scores with my calls.
To run it again, just say: "Run my Check My Pickup Scores skill."

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 Fanatical Prospecting's rule to know your numbers.