LinkedIn Reply Handling Prompt
A read of what the reply actually means and three responses at different levels of directness, including the one that ends it politely. The hardest message to write is the second one, after a reply that is neither yes nor no. This prompt separates what they said from what they meant, names the objection type, and writes the response — including, when the thread is dead, the message that closes it without pretending otherwise.
You are a senior seller reading a LinkedIn thread. You are good at the distinction between what someone wrote and what they meant, and you would rather close a dead thread than keep it alive on hope.
## Context
Me: {{ROLE}} at {{COMPANY}}. What we sell: {{PRODUCT}}
Proof I can cite: {{PROOF}}
My objective: {{OBJECTIVE}}
Tone: {{TONE}}
## The thread
[PASTE THE WHOLE THREAD, oldest message first, with dates. Include my messages, not only theirs.]
## Method
1. Classify the reply. One of: interested, deferring, delegating, objecting on price, objecting on fit, objecting on timing, objecting on incumbent, testing, or refusing.
2. State what they said and, separately, what it usually means. Deferring replies ("circle back in Q3") and delegating replies ("talk to my colleague") are the two most often misread — one is usually a soft no and one is usually a real opening, and which is which depends on whether they named a person.
3. Decide whether this thread is worth continuing. Be willing to say no. Include the evidence.
4. Write three responses:
- Advance: moves toward the objective by the smallest possible step.
- Clarify: asks the one question that would tell you whether this is real.
- Close: ends the thread gracefully, leaving the door open with a named condition. This is the one people never write and the one that most often gets a reply.
5. Recommend one and say why.
6. Note what this reply reveals about my earlier message, so the next first-touch is better.
## Constraints
- Do not argue with the objection. Acknowledge, then either add information or ask a question.
- Do not offer a discount. It is not in scope and it converts a fit problem into a price problem.
- Under 500 characters per response. Print the counts.
- Never imply a previous conversation, agreement or introduction that is not in the thread.
- If they said no clearly, the only acceptable output is a clean acknowledgement and an exit. Do not write a rebuttal.
## Output format
### What they said / what it means
Two short paragraphs, clearly separated.
### Classification
The type, and how confident you are.
### Worth continuing?
Yes or no, with the evidence.
### Response A — advance
Text, then (n characters).
### Response B — clarify
Text, then (n characters).
### Response C — close
Text, then (n characters).
### Recommended
Which, and why in one sentence.
### What this tells me about my first message
One or two lines.- You got "send me some info" and know that sending info ends it.
- You got a price objection before they know what it does.
- They replied warmly two weeks ago and then stopped.
- You are training someone and want the reasoning written down, not just the reply.
- Paste the whole thread including your own messages. The model cannot judge a reply without seeing what it was replying to.
- Keep the dates. "Replied enthusiastically and then went quiet for three weeks" is a different situation from the same words sent yesterday.
- Take the "worth continuing?" verdict seriously. Most pipelines are inflated by threads that were closed by the prospect and reopened by the seller.
- Collect the "what this tells me" lines. After twenty of them you have an objection list worth putting in OBJECTIONS.md.
The output schema it demands
The single highest-leverage part of any prompt. Without a declared shape you get prose you then have to reformat; with one, you get sections you can act on — and you can tell at a glance when the model skipped something.
### What they said / what it means ### Classification ### Worth continuing? ### Response A — advance ### Response B — clarify ### Response C — close ### Recommended ### What this tells me about my first message
What it looks like filled in
Illustrative, not a promise — this is the shape of a good answer rather than a guarantee of one. The scenario is the same fictional company used across the whole toolkit.
Thread: sent a personalised note about her long close. She replied four days later: "Thanks for this — send me some info and I'll take a look when things calm down."
What they said / what it means — literally: send information. Usually: a polite exit — though the specific timeframe ("when things calm down") is slightly more open than a flat brush-off would be.
Classification — deferring, medium confidence this one is real rather than a soft no.
Response B — clarify — "Happy to — one question first so I send the right thing: is it the entity-count side of the close that's the pain, or something else? Two very different one-pagers."
Recommended — B. It tests whether "send info" was genuine without spending a full pitch if it was not.
Do not do these
Each one is a real failure mode, not a disclaimer.
- Sending "info" when asked for info. It is the standard polite exit; answering it literally completes the exit.
- Treating a clear no as an objection to handle. It costs the relationship and occasionally the account.
Attach these once and stop retyping your business
This prompt works pasted into a blank chat. It works considerably better inside a Project with these attached, because then the ICP, the product and the limits are already in context and every answer inherits them.
OBJECTIONS.md
The assistant answers the objection you actually get, in the way that has worked, instead of improvising.
Free downloadOUTREACH.md
Every message an assistant writes comes out at the right length, in the right shape, with the openers you hate already banned.
Free downloadPRODUCT.md
Messages stop describing features and start describing what changes for the person receiving them.
Free downloadThis is one step in a sequence
The output of the step before is the input to this one, and this one's output is the next one's input. That is the whole reason these are not independent pages.
The reference behind this prompt
An InMail credit is returned when the recipient replies within 90 days. Replies are literally the currency of the allowance.
If the objection you keep hearing is your own tooling budget, the prices are here, dated.
Written to a fixed structure and run through Claude and ChatGPT to confirm each one returns the output format it declares. That is not the same as a measured reply-rate benchmark, and nothing here claims one — if you run these at volume and have numbers, we want them.
Take the whole folder instead
This prompt, the files it expects and the rest of the sequence, as a zip you can drop into a Claude Project, a Cursor workspace or an agent repo. Free, no account.