A LinkedIn lead magnet post only works if everyone who asked gets the link, including people you are not connected to. Pick a keyword nobody types by accident and match it as a whole word. Message your connections directly. Everyone else needs a connection request first, and that queue is slow: HeyReach caps a sender at 200 a week.
Disclosure: Signl is ours. The last section shows how Signl handles these posts. Everything before it works with any tool, or by hand.
Key facts
- LinkedIn lets you message 1st-degree connections for free. Other members only through Open Profile, a shared group, or InMail (LinkedIn Help, read 4 October 2026).
- LinkedIn publishes no weekly invitation number. It restricts accounts that send many invitations quickly, or whose invitations are ignored or left pending (LinkedIn Help).
- HeyReach caps each sender at 200 connection requests a week, with a default of 25 a day (HeyReach Help, read 4 October 2026).
- Signl checks a lead magnet post every 4 hours for its first 7 days, then daily.
What actually goes wrong with "comment GUIDE and I'll send it"
The guides ranking for this query on 4 October 2026 mostly cover the same ground: write a hook, offer a free asset, ask for a keyword, deliver by DM. One delivery tool promises the DM "the moment someone comments" (Lead Magnet Pilot, read 4 October 2026). Another, LinkMagnet, says plainly that unconnected commenters get the resource only after they accept a connection.
None of the pages we read on 4 October 2026 do the arithmetic on the people you are not connected to. A post that pulls 400 askers from outside your network does not need 400 messages. It needs 400 connection requests, each accepted, then 400 messages. At HeyReach's weekly cap, that is two weeks of one sender's requests before anyone else gets one. Most of the failure lives in that gap.
The playbook, step by step
1. Pick a keyword nobody writes by accident
The keyword is a filter. Its job is to separate "send it to me" from "great post". A common word fails that job, because people write it anyway.
GUIDE on a post about a guide is the classic mistake. "Great guide, thanks" contains GUIDE as a whole word. So does "is there a guide for agencies?" Both people get counted as asking. A made-up token is safer: PRICING26, OPSKIT, a product codename. We think an odd word costs you nothing, because people copy whatever the post tells them to type.
2. Match it as a whole word, in any case
Substring matching counts the wrong people. With a keyword of GUIDE, "misguided" and "guides" are not asks. But "guide!", "#GUIDE" and "Guide please" are.
Two details catch people out:
- Bold Unicode. Some people paste the keyword in styled letters (๐๐จ๐๐๐). Fold the text to plain characters before matching, or you miss them.
- The second comment. Someone writes "Interesting" on Monday and "GUIDE" on Tuesday. If you only read each person once, you never see the ask.
3. Decide: keyword, or everyone who comments
Dropping the keyword and promising the asset to every commenter is a legitimate choice. You stop missing the people who write "send please" instead of the keyword. You also deliver to people who only wanted to say something. Our view: use a keyword when the asset is the point of the post, and skip it when the post is the point and the asset is a bonus.
Reactions are never asks. A like carries no words, so nobody asked for anything.
4. Write the delivery message, and leave the pitch out
The first message carries the link and nothing else. Name the thing they asked for, use their first name, and include the URL. A pitch in the delivery message turns a kept promise into a cold message they were tricked into. Save the conversation for a follow-up (step 8).
5. Build the campaign in two branches
Connected and not-connected people need different first steps. HeyReach sequences have an "If connected" condition that sends already-connected leads straight to messaging (HeyReach Help). The other branch is a connection request, then the delivery message once they accept.
Expandi's help says a connector campaign is for 2nd and 3rd-degree people only. It adds that first-degree connections cannot receive another connection request (Expandi Help, read 4 October 2026). Before relying on Expandi for delivery, check how your campaign treats someone you are already connected to.
6. Size the queue before you post
This is the step nobody does. The numbers that set it, as of 4 October 2026:
| Limit | Value | Source |
|---|---|---|
| LinkedIn invitations per week | Not published | LinkedIn Help |
| HeyReach connection requests per sender, weekly | 200 | HeyReach Help |
| HeyReach default daily requests (warmed-up account) | 25, up to 40 | Same page |
| Expandi recommended limits | Not readable without a login on 4 October 2026 | Expandi Help |
| LinkedIn restriction length | "Most" lift within one week | LinkedIn Help |
HeyReach applies its limits per sender, across every campaign that sender runs. A sender already prospecting at full volume has no room left for a magnet. If you expect several hundred askers from outside your network, either add senders or say in the post that delivery takes a few days. Accepting requests is also up to them. Some never accept, and those people never get the link by DM.
7. Keep reading the post after day one
Comments keep arriving after the first day. If your tool reads the post once, or only while the browser tab is open, everyone after that gets nothing. Our view: a late asker who gets nothing is worse than no post, because they asked in public. Check the post several times a day while it is fresh, and at least daily for weeks after.
8. Deliver to everyone, follow up with the fit
You promised the asset to everyone who asked, so qualification does not gate delivery. It gates the follow-up. A founder at a company in your market who asked for your pricing playbook is a conversation. A student who asked for it is a kept promise and nothing more. Score each asker against your ICP after delivery. Then put only the ones who fit into a second-touch sequence that answers what the asset was about. How to turn LinkedIn comments into leads covers the qualification half in detail.
What kills lead magnet posts
Engagement pods. Pod members comment the keyword to lift reach, not to read anything. They inflate the ask count with people who will never buy. LinkedIn's Professional Community Policies say: "don't agree with others ahead of time to like or re-share each other's content." We would not run a magnet through a pod.
Commenters without the keyword. "Interested!", "Send please", "+1". You either widen the net (step 3) or reply to them in the thread with the keyword. Their next comment then counts.
The post stops being checked. The typical cause is a tool that watched the post for a day, or a browser extension on a laptop that closed. The askers after that point are your most visible failures.
Your own replies. Writing "Sent!" under each comment is fine. A tool that counts the post's author as an asker is not.
Automation risk. LinkedIn's User Agreement section 8.2 prohibits "unauthorized automated methods" to send messages, among other things. HeyReach and Expandi are third-party tools that send from your account. That account risk is yours to weigh, and it applies to every magnet delivery tool, ours included.
How Signl does it
Signl watches the post and decides who asked. It hands those people to your HeyReach or Expandi campaign, which sends the message. Signl does not send LinkedIn messages itself.
When you add a source, tick This post is a lead magnet. The settings are:
| Setting | What it does |
|---|---|
| Keyword | Matched as a whole word, any case, after folding styled Unicode to plain text. Empty means everyone who comments. An emoji keyword falls back to a plain substring test. |
| What you promised | The asset's name, used in the message ("the pricing playbook"). |
| Link to it | Must be an https URL. |
| Message they get | Placeholders {first_name}, {asset_name}, {asset_url}. It must include {asset_url}. Empty uses the default. |
| Delivery | HeyReach (pick a campaign) or Expandi (the campaign whose webhook you connected). |
| Send it automatically after every check | Creates an autopilot rule bound to this post. |
What happens next:
- Checks. The post is read every 4 hours for 7 days, then daily. Lead magnet posts go first in each run.
- Matching. Only written comments and replies can ask. Someone who commented without the keyword and later commented with it is moved to asked.
- No filtering of askers. Everyone who asked lands in To contact, whatever their ICP fit. The only person never counted is the post's own author. Askers are still scored for fit, so you can sort them for step 8.
- Delivery. Put
{{signl_dm}}in the campaign's first message. Signl fills it with the rendered message for each person. Everyone from that post who asked joins the campaign, oldest ask first. - Pace. The rule the form creates sends up to 50 people a day, and an account's autopilot rules share 500 a day in total. Autopilot only considers people found in the last 14 days. Anyone older is sent by hand from To contact.
- Replies. Replies come back from HeyReach, so the follow-up happens in one place.
One limit to plan around: each check reads a post newest first, up to 500 comments, the same on every plan. A post that draws more comments than that between two four-hourly checks can leave the earliest of that burst unread.
Cost. Nothing per asker. As of 24 October 2026, pricing is by the sources you watch, with no credits: Self-serve is $99 a month for 25 sources and unlimited seats, or $79 billed yearly. Each asker's verified work email is included within a monthly fair use of 500, and if that runs out, lookups wait until the 1st of the next month. The detail is in plans and pricing, and why only verified emails are worth paying for is in the bounce tax.
Data. Signl collects public posts, comments and replies, plus reactions on the sources set to collect them. It never logs in to LinkedIn, deletes everything 90 days after it was collected, and never collects anyone who objects again. The lawful-basis question is covered in GDPR and B2B prospecting. This is not legal advice.
Moving off Trigify? What to move before 22 October covers the export and the switch.
Two neighbours worth reading: how to message someone who commented on a post for the people who asked but never replied to the DM, and sending engagement leads to HubSpot and Clay for where the asset and keyword fields land downstream.
A lead magnet post trades qualification for volume. For the post shapes that do the opposite, and the closing question that makes a buyer describe their situation instead of typing a keyword, see LinkedIn posts that bring buyers into the comments.
