DM Sequence Framework
After reading this guide, you will have a documented DM sequence framework with message types, timing gaps, and conditional branching logic configured per ICP tier in your LinkedIn automation tool.
TL;DR
4 message types, 3 sequence formats: the complete framework
A DM sequence framework defines how many messages you send after a LinkedIn connection accepts, what each message does, and how long to wait between steps. Most sequences fail because follow-up messages restate the opener instead of advancing toward a decision.
Message types, timing gaps, 3 sequence templates (short, standard, signal-based), conditional branching for non-accepts, and which tools handle if/then logic natively.
The Framework
Three DM sequence scenarios at a glance
| Dimension | Short (2-3 DMs) | Standard (4-5 DMs) | Signal-Based (3 DMs) |
|---|---|---|---|
| Volume fit | 50+ new connections/week | 15 to 50 connections/week | 5 to 20 signal-qualified/week |
| Message 1 | ICP opener + direct ask | ICP opener + shared context | Signal reference + opener |
| Message 2 | Bump or breakup | Value-add resource or insight | Value-add or direct ask |
| Message 3 | Optional breakup | Angle shift (different pain) | Breakup |
| Message 4 | - | Bump | - |
| Message 5 | - | Breakup | - |
| Gap between steps | 2 to 3 days | 3 to 5 days | 2 to 4 days |
| Best fit | High-volume, lower ACV | Mid-market, defined ICP | Intent-triggered, named accounts |
Message Types
4 DM types: what each step must accomplish
Message 1 names a context: a shared ICP signal, a recent profile trigger, or an explicit reason the connection makes sense right now. It does not pitch. A note already sent in the connection request reduces the obligation on message 1 since partial context is set.
Message 2 introduces something new: a case study, a data point, or a different problem framing. Restating message 1 with different wording is not a value-add. If both messages name the same problem and make the same ask, the sequence stalls.
- Angle shift (message 3 in a standard sequence)
Approach from a different angle: a second use case, a different stakeholder problem, or a business outcome instead of a feature. The prospect who did not respond to the first frame may respond to a second one.
- Bump (message 4)
One or two lines that resurface the thread without adding new content: "Wanted to make sure this landed" followed by the single most relevant sentence from a previous message. Bumps generate meaningful late replies from contacts who read earlier messages without responding.
- Breakup (final message)
Closes the sequence and often generates the highest reply rate of any step. Frame non-response as understandable and leave a door open: guilt without a path to re-engagement underperforms.
Each message should be auditable on its own: if the previous message did not exist, would this one still make sense? If no, it is restating, not advancing.
Timing and Volume
Timing gaps and daily DM caps: 80 to 100 combined actions per day
Standard gap between DM steps: 3 to 5 days for mid-market. Compressing to 24-hour gaps reads as aggressive and concentrates daily action counts above LinkedIn's automation detection threshold.
Most tools enforce a safe ceiling of 100 to 150 follow-up messages per account per day. The practical safe limit running both connections and DMs is 80 to 100 combined actions daily, with connection requests kept below 20 per day.
Spread sequences across warmed accounts using a tool with multi-sender rotation. More accounts sending less per account, not one account sending more.
Conditional Branching
Non-accept after 7 days: 2 actions that recover the contact
A DM sequence only fires after a connection request is accepted. For requests unaccepted after 7 to 10 days: export to cold email if you have a verified address, or withdraw the request and re-queue the contact after 60 days.
Tools with native if/then logic (La Growth Machine, Skylead, Expandi) handle this branch automatically. Without conditional logic, manually audit pending requests weekly and export non-accepts to cold email.
If accepted within 7 days, start the DM sequence at message 1. If not accepted, export to cold email or withdraw and re-queue at 60 days.
Recommended Tools
3 tools with native if/then branching: La Growth Machine, Expandi, Dripify
La Growth Machine and Expandi handle if/then branching natively; La Growth Machine also triggers email steps from the non-accept branch. Dripify is the lowest-cost entry point for LinkedIn-only sequences without a multichannel layer.
Common Questions
5 questions on sequence length, timing, and non-response
Four to five DMs covers most mid-market B2B sequences. Two to three works for high-volume broad lists; beyond five rarely improves reply rates.
No. Connection request notes should establish why the connection makes sense for the prospect. The pitch belongs in the DM sequence post-acceptance.
Three to five days between steps is standard for mid-market; two to three days works for high-volume short sequences. Gaps under 48 hours read as aggressive and increase spam flag risk.
Three messages over 6 to 10 days: message 1 references the specific signal (job change, funding, post), message 2 adds a relevant resource, message 3 is a breakup. Signal sequences are shorter because personalization in message 1 justifies a direct ask earlier.
After the breakup, tag as non-respondent and move to cold email if you have a verified address. Wait at least 60 days before re-engaging on LinkedIn.
Sequence built. Now pick the right LinkedIn automation tool to run it.
Compare LinkedIn automation platforms on conditional logic, safety controls, and pricing before configuring your DM sequence.