LinkedIn DMs work when they reach the right person at the right time with the right message. They fail — and trigger account restrictions — when operators copy-paste the same template across hundreds of recipients. The platform's Professional Community Policies prohibit automation explicitly; enforcement is pattern-based, and message similarity is one of the fastest ban triggers in 2026.

This post is 8 LinkedIn DM templates organized by intent type, with the structural rules behind each one and the five mistakes that get accounts flagged. The templates are scaffolding, not copy-paste blanks — every DM has to reference something specific the recipient wrote, posted, or accomplished, or LinkedIn's message-similarity classifier trips within a handful of sends.

Key takeaways

• LinkedIn DMs that reference specific posts or context run 12–18% reply rates on warmed accounts; templated DMs run under 3% and get restricted within weeks.

• Four intent types map to four template families: direct ask (highest), competitor frustration, problem statement, and engagement (long-game).

• Three structural rules apply to every LinkedIn DM: reference something specific, keep it under 3 sentences, no link in the first message.

• Account history matters as much as message quality — the 7-day warmup playbook is non-negotiable for new or dormant accounts.

• Templates 1–3 are highest-converting; templates 7–8 are situational long-game plays for high-fit prospects not currently in market.

Why most LinkedIn DMs get ignored or trigger restrictions

LinkedIn's Professional Community Policies explicitly prohibit "automated tools to send connection requests or messages." Enforcement is pattern-based: the message-similarity classifier flags accounts whose DMs cluster at 60%+ structural similarity, even with first-name merge fields swapped. The LinkedIn Transparency Center reports automated enforcement up 31% YoY — detection is faster and stricter every quarter.

The DMs that work share three qualities: they reference something specific from the recipient (post, comment, job change, mutual connection, company news), they're under 3 sentences, and they don't include a link in the first message. Get any one of those wrong and reply rates collapse to under 3%. Get all three right and the channel runs at 8–18% reply rates on warmed accounts — the highest single-channel reply rate available to solo founders in 2026.

The templates below are organized by the 4 intent types we use to classify B2B signals: direct ask, competitive (frustration with named competitor), problem (problem statement, no tool named), and engagement (long-game relationship building).

The 3 structural rules every LinkedIn DM follows

1. Reference something specific. Their post, their comment, their job change, their company news, a mutual connection. Not "I came across your profile" — that triggers the spam classifier and reads like every cold DM the recipient deletes daily.

2. Keep it under 3 sentences. Long DMs feel like sequences. Short ones feel like a real person typing. Three sentences is the sweet spot for warmth without bloat.

3. No link in the first message. LinkedIn weights links in early messages as a spam signal. Mention the product or page by name; let them ask. "Happy to share more if useful" converts better than "check this out: [link]."

Meta-rule: write like a peer who happened to find them, not a sales rep running a campaign. Every template below is scaffolding to be adapted to the specific person — never sent verbatim.

Direct-ask templates (intent score 9–10)

Direct asks on LinkedIn are rarer than on Reddit but higher-converting when they happen. Common forms: someone posts asking their network for a tool recommendation, comments under a tool review asking for alternatives, or rants about a competitor and explicitly asks what to switch to.

Template 1 — Network recommendation request

"Saw your post asking for [category] recommendations — hit the same wall last year on [their constraint]. Built [your product] specifically for that case (different approach: [one line]). Worth a look?"

Use when: prospect explicitly asks their LinkedIn network for a product recommendation in your category. Highest-converting LinkedIn DM template by a wide margin. Reply within 2–4 hours; LinkedIn posts have longer half-life than Reddit but engagement still decays.

Template 2 — Comment under tool review

"Read your comment under [post] — the [specific limitation] thing is real. We built [your product] partly because of that. Different model: [one-line]. Curious if you've found something that works."

Use when: prospect comments under a tool review or LinkedIn newsletter rant, asking for alternatives. The opening shows you actually read their comment (not just their profile); the close invites conversation.

Competitor-frustration templates (intent score 7–9)

Competitor frustration is the most common high-intent signal on LinkedIn. Founders rant about pricing, missing features, support, or contract terms publicly because LinkedIn rewards opinion content with reach.

Template 3 — Pricing rant

"Read your post about [competitor] pricing — same thing pushed us off it last year. Different approach with [your product]: [one-line pricing model]. No pitch, just curious where you ended up."

Use when: prospect publicly criticizes a competitor's pricing. Acknowledges shared pain (without piling on the competitor); offers your alternative; closes with curiosity, not CTA.

Template 4 — Feature gap rant

"Saw your post about [competitor] missing [the limitation]. We built [your product] specifically around that gap. Worth a look at [one-line approach]?"

Use when: rant focuses on a specific missing feature your product has. Direct, specific, and short — acknowledges the gap, offers your solution, asks a question.

Problem-statement templates (intent score 6–8)

Problem statements on LinkedIn are buyers describing a pain without naming a tool. Common forms: founder posts about pipeline struggles, marketer posts about reply rate decline, agency owner posts about delivery problems.

Template 5 — Pipeline / outbound problem

"Saw your post about [their problem] — same boat last year, spent 6 months on [their existing approach] before realizing [the actual diagnosis]. Different angle that worked: [your approach by name]. Happy to walk through it."

Use when: founder describes pipeline, sales, or outbound problem. Establishes shared context; offers experience over pitch.

Template 6 — "I tried X and it's not working"

"Read your post about [their failed approach]. Common pattern — [the actual reason it failed]. Switched to [your approach] last year, different math: [one-line outcome]. What's been your biggest blocker specifically?"

Use when: prospect describes something they tried that didn't work. Diagnoses why; offers your alternative; closes with a clarifying question.

Engagement templates (intent score 5–7)

Engagement signals are LinkedIn posts adjacent to your category but not direct asks. Long-game pipeline; reply only if you can add real value.

Template 7 — Tangential post engagement

"Reading your post about [adjacent topic]. Slightly off-topic but related — we hit a similar problem on the [your category] side and ended up at [your approach]. Curious what you tried."

Use when: signal is adjacent to your category. Be honest about the off-topic nature — force-fitting reads as spam.

Template 8 — Long-game relationship opener

"Reading your work on [their general topic] — not pitching anything. We work in [adjacent space] and your post resonated. If you ever look at [your category] for [the related use case], we're around."

Use when: high-fit prospect not currently in market. Plants a flag; opens a relationship for when they are in market.

When to use which template

Reply rates assume the structural rules are followed, the account is properly warmed, and acceptance rate is above 30%. Bad targeting drops every template to zero regardless of message quality.

The 5 mistakes that get accounts restricted

1. Sending the same template across multiple recipients. LinkedIn's message-similarity classifier flags 60%+ structural similarity, even with names swapped.

2. Including a link in the first message. Spam classifier weight goes up sharply.

3. Long DMs. Anything over 4–5 sentences feels like a sequence to the recipient and the classifier.

4. Skipping warmup. New or dormant accounts sending DMs with no behavioral history are the fastest ban path. The 7-day warmup playbook is mandatory.

5. Ignoring acceptance rate. Below 30% acceptance LinkedIn assumes you're spamming and throttles you. Tighten targeting before sending more.

The full discipline is in How to DM on LinkedIn without getting banned. The templates above plug into that playbook; the playbook keeps your account alive long enough for the templates to compound.

How automation scales LinkedIn DMs without losing personalization

Running 8 template families against high-intent LinkedIn signals by hand maxes out around 25–30 DMs/day per account, roughly 3–4 hours/week. Beyond that, classifying intent, drafting personalized DMs, and respecting volume caps becomes the bottleneck.

repco's cross-platform detection automates the whole loop: watches LinkedIn every 2–4 hours, classifies intent on a 1–10 scale via Claude Sonnet 4.6, picks the right template family based on intent_type, drafts a 3-sentence DM that references the specific signal, and sends from your account with built-in warmup, behavioral noise, and volume caps. Same templates, but customized per-signal and run 24/7. How repco compares to Phantombuster covers the LinkedIn-specific automation choice.

If you're not yet at scale, the manual playbook with these 8 templates beats any cold email setup at any volume. Start there.

Frequently asked questions

Can I copy-paste these LinkedIn templates verbatim?

No — the templates are scaffolding. Every DM has to reference the specific post, comment, or context of the recipient. Copy-paste verbatim trips the message-similarity classifier within a handful of sends and accounts get restricted within 2–6 weeks. Adapt every template to the specific signal.

Should I include a link in the first LinkedIn DM?

No. LinkedIn weights links in early messages as a spam signal. Mention the product or page by name; offer to share the URL if the recipient asks. "Happy to share more if useful" converts better than "check it out: [link]."

How long should a LinkedIn DM be?

Under 3 sentences for cold outreach. Long DMs feel like sequences — the recipient skims them, the classifier groups them with automation patterns. Three sentences is enough to reference something specific, offer your angle, and ask a question.

What's the difference between DMing and connection requesting on LinkedIn?

Connection requests have stricter limits and acceptance rates determine your account-level restrictions. DMs to existing connections (1st-degree) are higher-trust but lower-volume because you have to build the network first. The right pattern: send a personalized connection request, send a DM after acceptance. The full LinkedIn discipline rules cover the volume caps for both.

Bottom line

LinkedIn DMs work when they reference something specific, stay under 3 sentences, and don't include a link in the first message. The 8 templates above cover the four intent types you'll see most often — direct ask, competitor frustration, problem statement, engagement — with the structural rules each follows.

Use them as scaffolding, not blanks. Personalize every send. Respect the volume caps. The channel runs 8–18% reply rates on properly scored signals — multiples of cold email — but only if you treat your account as a long-term asset.

The full LinkedIn ban-prevention playbook covers warmup and discipline. The 1–10 intent score framework covers how to classify before writing. The complete 2026 outbound guide covers where LinkedIn fits in the broader pipeline.

About the author

Kamil is the founder of repco.ai — the AI sales rep that finds buyers publicly asking for products like yours on Reddit and LinkedIn. 15 years across marketing and sales, building and running companies in industrial, IT, investments, and real estate. Serial founder; building repco from the gap he kept hitting himself — outbound channels that work for solo founders and small teams, not enterprise sales orgs. Wrote and tested every template above across hundreds of LinkedIn DMs before encoding them into repco's automated drafting.