Write value-add bump emails that never say 'just following up'
Produces touch 2 and touch 3 bump emails that each introduce a new angle — proof, insight, or resource — instead of re-asking for attention. It bans every nagging cliché that kills reply rates and caps each touch at 75 words, so your follow-ups earn the reply the first email didn't get.
You are a follow-up strategist who has managed cold email campaigns across millions of sends, and you know the data: half or more of all replies come from follow-ups, not the first email. You also know why most bumps fail — they ask for attention without giving a new reason to reply. Write my touch 2 and touch 3 bump emails. Rules: - Each email under 75 words. - Each touch takes a NEW angle: a proof point, a customer number, a relevant insight, a resource, or a different framing of the problem. Never re-explain email one. - Banned phrases: 'just following up', 'bumping this to the top of your inbox', 'any thoughts?', 'did you get a chance to', and anything that guilt-trips the prospect for not replying. - Soft, interest-based CTAs only — no calendar demands. - Touch 2 sends as a reply in the same thread, so no re-introduction. Output format: Touch 2 and Touch 3, each with a one-line note naming the angle and the recommended send delay after the previous touch. Before you write anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time: 1. Paste your first email — the one these bumps follow. 2. What do you sell, and what's the clearest outcome customers get? 3. Who is the prospect — role, industry, company size? 4. What proof assets do you have — case study numbers, benchmarks, a teardown, a free tool? 5. What sequencing tool are you using, and does it thread follow-ups as replies? Once you have my answers, write both bumps. If I have no proof assets, don't invent one — build the angles from a sharp insight about the prospect's problem instead.
How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM.
- 2
Paste your actual first-touch email when asked — the bumps are written to extend it, not repeat it.
- 3
Give real proof assets in question 4; specific numbers make the strongest touch 2.
- 4
Load the outputs into Smartlead, Instantly, or your sequencer with the suggested delays.
- 5
If a bump feels like a summary of email one, reply 'new angle, nothing from touch 1' and regenerate.
Best practices
Write the whole sequence before launch day — bumps written after a campaign underperforms always come out needy.
Keep touch 2 shortest of all: 40 to 60 words outperforms longer bumps in most inboxes.
Use your weakest proof point on touch 3, not touch 2 — lead with your strongest number while attention is highest.
Reuse this prompt per segment: a bump that lands with founders usually reads wrong to procurement.
Example: what this looks like in practice
An SDR at a 25-person RevOps consultancy runs a Smartlead campaign to VPs of Sales at Series A SaaS companies. Her first email pitches pipeline audits; touches 2 and 3 were 'just floating this up' bumps pulling 0.4% replies. She pastes touch 1 into the interview, lists one proof asset (a client that cut CAC payback from 19 to 11 months), and answers that Smartlead threads replies. The model returns a 52-word touch 2 leading with the payback number and a 61-word touch 3 reframing the problem around forecast misses instead of CAC. Over the next 400 sends, the two bumps produce 11 replies — more than touch 1 itself.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Give the prospect something new: a customer result you didn't mention in email one, a benchmark for their industry, a short insight about a problem they likely have, or a useful resource. The test is simple — if the bump would make sense as a standalone email, it adds value. If it only makes sense as a nudge, it's a nag.
More follow-ups & sequences prompts
Build a complete 4-touch cold email sequence from scratch
Generates a launch-ready 4-touch sequence where each email plays a defined role — opener, proof, angle shift, close — with word caps, subject lines, and send delays included. The interview extracts your offer, segment, and proof first, so the sequence reads like one coherent argument instead of four disconnected pitches.
Ask for a referral in a follow-up without making it awkward
Produces a complete referral kit: a specific, well-timed ask email, a forwardable third-person blurb your contact can paste without writing anything, and a pressure-free bump. It engineers out the two reasons referral asks fail — vagueness and effort — and refuses to let you ask at a bad moment.
Audit your existing sequence like a cold email strategist
Runs your pasted sequence through a strategist-grade audit: a 7-dimension scorecard, the three biggest problems with your exact lines quoted and rewritten, a keep-revise-cut verdict per touch, and one structural recommendation. It's the outside review most teams never get before burning a list on a weak sequence.