Write a 3-touch follow-up sequence for a lead that went quiet, each touch a different angle

What this does

You summarize where things stand. The AI drafts three follow-ups: one with a question, one with something useful, and a polite close-out that gives them an easy way to say no or not now.

Who it’s for

Reps and owners with someone who stopped replying after interest or a proposal. Not for people who said no or asked you to stop; don’t contact them.

What you need

• What you sent or discussed before (2-3 sentences)
• Their situation, in your own notes
• A useful point you can back up for touch 2 (or “none”)
• What you want them to do next
• Days between touches
• Tool: Any AI chat tool (for example ChatGPT, Claude, Gemini, or Grok), Your email or CRM (manual send)
• Cost: $0 with a free AI chat tier (limits change)
• Setup: 3 minutes (estimate)

Steps

1. Write 2-3 sentences on what happened so far. Don’t paste the whole email thread unless your policy allows it.
2. Pick spacing for touches. The first number is days from today for touch 1; each later number is days after the previous touch (3, 5, 7 means send on day 3, day 8, and day 15). Choose what fits your sales cycle. Also decide on one true, useful point you can back up for touch 2, or write “none.”
3. Run the prompt. Read all three.
4. Remove or fix anything that isn’t true or that adds pressure you don’t mean.
5. Add your opt-out line and your company’s postal address (the AI reminds you but does not write them). Schedule or send each yourself. Stop the sequence immediately if they reply or opt out.

The prompt

Write a 3-touch follow-up sequence for a salesperson. The prospect went quiet. Use only the facts I give you. Do not invent deadlines, discounts, or news.

NO-ADDITIONS RULE: Do not add facts, tips, statistics, scarcity or urgency, promises, references or testimonials, or the gender of any person that I did not supply.

What happened so far: [2-3 SENTENCES]
Prospect's role: [ROLE]
What they were interested in: [TOPIC]
A useful point I can back up (or "none"): [USEFUL POINT]
What I want them to do next: [ONE ACTION]
Days between touches: [e.g. 3, 5, 7]. This means touch 1 goes out that many days from today, then each later touch goes out that many days after the previous one. Example: 3, 5, 7 = send on day 3, day 8, and day 15. Label each touch with its send day.
My voice: [e.g. direct, warm, short]

Write three emails, each under 80 words, each with a different angle:
- Touch 1: A specific question that is easy to answer in one line.
- Touch 2: Use ONLY the point I gave in [USEFUL POINT]. If I wrote "none," the only content you write for the useful point is this placeholder: [ADD A REAL USEFUL POINT]. Do not write a tip, checklist item, or fact of your own.
- Touch 3: A respectful close-out: say I won't keep following up, offer to reconnect later, and make it easy to reply "not now" or "no."

Rules: No "just checking in," "circling back," or "touching base." No guilt, no fake urgency, no pretending we spoke if we didn't. Subject lines are honest and under 6 words. After each, add: "If they reply with X, my next move is Y" (one line).

At the very end add this line exactly:
"Compliance reminder for you (not part of any email): before sending, add your opt-out line and your company's postal address. I have not written them; you must supply them."

Sample input (fake data)

All names and companies are fictional. This is the input for run 4 (the output below), on the current prompt. Runs 5 and 6 used different inputs; see the test log.

What happened: Demo on Oct 1 for a 30-person landscaping business owner; they asked for pricing and I sent a pricing sheet on Oct 2. Nothing since.
Role: Owner
Interest: Route planning for crews
Useful point I can back up: Our pricing sheet shows the plan is $149 a month for up to 10 crews
Next action: Book a 15-minute call
Days between touches: 2, 4, 6
Voice: direct, friendly, short

Sample output

Real output from run 4 on 2026-10-14, using the current prompt, from an unspecified AI assistant (Grok Bot agent session; model name not shown). Copied as produced. Nothing was fixed after the fact; the problems we found in it are listed under “How it went”.

**Touch 1 (send day 2, which is 2 days from today)**
Subject: One question on pricing
Hi,

After the Oct 1 demo I sent the pricing sheet on Oct 2. One question: did anything on it raise a question for you?

A one-line answer is plenty.

If they reply with a question, my next move is to answer it in one email and offer a 15-minute call.

**Touch 2 (send day 6, 4 days after touch 1)**
Subject: Plan size and price
Hi,

One detail from the pricing sheet in case it helps: the plan is $149 a month for up to 10 crews.

If that's worth a closer look, I'm glad to book 15 minutes.

If they reply with how many crews they run, my next move is to confirm whether the plan fits and offer the call.

**Touch 3 (send day 12, 6 days after touch 2)**
Subject: Closing the loop
Hi,

I haven't heard back, so I won't keep following up. If route planning becomes a priority later, I'm happy to pick this up again.

If now isn't the time, a quick "not now" or "no" is completely fine.

If they reply "not now," my next move is to note a date they pick and stop emailing until then.

Compliance reminder for you (not part of any email): before sending, add your opt-out line and your company's postal address. I have not written them; you must supply them.

Check before you send

• Touch 2’s ‘useful point’ is real and you can back it up. No placeholder left in.
• No deadlines, discounts, or claims you haven’t approved.
• The person hasn’t said no or unsubscribed.
• Subject lines are honest and match the content.
• Email footer includes what your company/region requires: your opt-out line and postal address, which the AI does not write.

How it went

• Test date, tool, model: 2026-10-14. Unspecified AI assistant (Grok Bot agent session; the model name is not shown to the agent). One tool only.
• Runs logged: 6, on 6 different fake inputs (see the test log). Runs 1-3 used the original prompt. Runs 4-6 are a second round on new fake inputs after the prompt was fixed, as listed below. Every run was done by the same AI assistant session that wrote the fake input, and in round 2 the same session also edited the prompt, so these are not blind or independent runs. A script counted words and characters against the prompt’s limits. No second person re-ran it from the written steps.
• Why the label is “Tried in an AI tool,” not “tested”: we reserve “tested” for plays with at least 3 logged runs in at least 2 different AI tools, and we ran one tool. We also have no independent reproduction of the steps and no timed manual baseline. The runs are real, but this is one tool’s behavior, and 3 runs per round is too few to say a problem is gone.
• Time saved: Not measured. We did not time a manual baseline or an assisted run.
• What we changed after round 1: Added: no-additions rule; a [USEFUL POINT] input (Touch 2 may use only that point, or the placeholder if ‘none’); ‘Days between touches’ defined as days from today for touch 1, then days after the previous touch, with each touch labeled with its send day; a closing compliance reminder (opt-out line, postal address).
• Round 2 (current prompt, runs 4-6). What worked:
  • No invented tip in touch 2 in any run. In the first round run 1 invented a point about how many times an invoice is touched. With ‘none’ (run 5) the model wrote only the placeholder.
  • Send-day labels matched the new definition in all 3 runs (2/6/12, 3/8/15, 4/8/18), so the ‘days between’ ambiguity is gone.
  • All 9 touches were under 80 words, subjects under 6 words, no banned phrases, every touch had the ‘If they reply’ line, and the compliance reminder was present each time.
• Round 2: what failed or needed fixing:
  • Run 5 touch 2 is now almost empty (16 words: a placeholder and one line). That is the intended behavior, but it means a rep with nothing to share gets a nearly blank email and edits take more than 2 minutes.
  • The rep has to supply a useful point or delete touch 2. The prompt doesn’t help find one.
  • The reminder is a single line at the end, below all three emails. It can be missed if a rep copies emails one at a time.
• Checked against “Check before you send”: Touch 2 point is real in runs 4 and 6 (supplied) and a placeholder in run 5; no invented deadlines or discounts; subjects honest; footer is still the rep’s job, with a reminder.
• Edge cases: No useful point (run 5): works as designed but yields a stub. Unusual spacing (4, 4, 10 in run 6) was labeled correctly.
• Round 1 (original prompt, runs 1-3): what failed, kept for reference:
  • Run 1 touch 2 wrote an invented ‘useful point’ (how many times an invoice is touched before it goes out) and also left the placeholder in. If the rep deletes only the placeholder, they send a tip nobody checked.
  • In all 3 runs touch 2 needs real content from the rep. That’s by design, but it means these drafts always need more than 2 minutes of work, so the edits column says No.
  • ‘Days between touches: 3, 5, 7’ was read as send days 3, 8, and 15 (a first touch on day 3, then gaps). The prompt doesn’t say whether the first number is the gap from today. A rep could read it differently.
• Not yet observed: how a different model or tool handles the same prompt (a second-tool run is still to do).

Watch-outs

Email rules apply (CAN-SPAM in the US: honest subject, identity, postal address, opt-out honored). Stop emailing anyone who opts out, and people in regions with stricter consent rules (GDPR/CASL) may need extra care. Not legal advice. Don’t auto-send; schedule only after you’ve approved each message.

See the Safe Use page: AI drafts, you decide, you send.

Variations

• Owner version: make Touch 1 a yes/no question about timing; keep Touch 3 to two lines.
• Manager version: have reps paste their sequence and ask the AI to point out where all three touches use the same angle.
• Text version: only for people who gave consent to be texted; see Safe Use.

Sources

None. Original wording. (Add any public framework or post used, with link and checked-on date.)

Retest

Re-check by: 2026-12-13 (60 days after the 2026-10-14 runs), or sooner when run in a second tool.

Test log

Changelog

• 2026-10-14: Draft written. Not run yet.
• 2026-10-14: Ran 3 times in one AI assistant on fake data. Filled in sample input and output, how it went, and the test log. Front matter set to runs_logged 3, last_tested 2026-10-14, retest_by 2026-12-13. The 2-tool bar for the word “tested” is not met, so the play is not labeled tested. The prompt text was not changed after the runs.
• 2026-10-14: Prompt revised after round 1. Added: no-additions rule; a [USEFUL POINT] input (Touch 2 may use only that point, or the placeholder if ‘none’); ‘Days between touches’ defined as days from today for touch 1, then days after the previous touch, with each touch labeled with its send day; a closing compliance reminder (opt-out line, postal address). Re-ran 3 fresh runs (4-6) on new fake inputs in the same AI assistant. Updated the sample input and output to run 4, the “How it went” section, and the test log (original runs 1-3 kept, marked). Front matter set to runs_logged 6, last_tested 2026-10-14, retest_by 2026-12-13. The 2-tool bar for the word “tested” is not met, so the play is not labeled tested.
• 2026-10-15: Published with the evidence label “Tried in an AI tool” (6 runs, 1 tool, fictional inputs). Earlier wording that implied more evidence than one tool’s runs was replaced with the label; the runs, results, and limits above are unchanged.