You identified a visitor. The alert dropped. Your AE was in a demo. By the time anyone looked, it was the next day, and the visit was too old to cold-call but too warm to drop into your standard cold outbound sequence.
This is the gap most sales teams fall into. Cold sequences are written for people who have never heard of you. Inbound sequences are written for people who filled out a form. Identified visitors sit in between: warmer than cold, less committed than inbound, and completely unserved by either sequence.
This post is the sequence that sits in that gap. A 5-touch warm outbound built for identified visitors. Real templates, real timing, and a hard rule on what you can reference versus what crosses into surveillance.
The answer up front: 5 touches over 11 days
| Touch | Day | Channel | Purpose |
|---|---|---|---|
| 1 | Day 0 (same day, inside Tier 1 SLA) | Relevance + permission to continue | |
| 2 | Day 2 | LinkedIn connect (no pitch) | Social proximity |
| 3 | Day 5 | Share a specific resource tied to their role | |
| 4 | Day 8 | LinkedIn message (post-connect) | soft ask, no ask |
| 5 | Day 11 | Direct ask, short, assumes close |
Total cadence: 11 days, 5 touches, 2 channels. Stop if they reply. Move to nurture if they don’t.
This is shorter than a cold sequence because they already know who you are. It is also gentler, because the reference to “you visited our site” can only be implicit. Never explicit.
The playbook
Step 1: Segment before you sequence
Not every identified visitor gets this sequence. Filter by:
- ICP fit: company size, industry, geography
- Role: decision-maker or known champion pattern
- Page intent: Tier 1 or Tier 2 only (per the follow-up SLA post)
- Suppression: not an existing customer, not a competitor, not a partner
Everyone outside those filters does not enter the warm sequence. They go to a Tier 3 nurture or are suppressed entirely.
Step 2: Write the templates with the “what I can reference” rule
Here is the rule that separates warm from creepy:
You can reference: role, company, the general topic or category they researched, industry context.
You cannot reference: the specific URL they visited, the exact time they visited, their return-visit count, how long they spent on a page, or anything that makes it obvious you were watching.
The reader’s instinct is “this email is relevant.” Not “this company was watching me.” Same signal, different phrasing.
Step 3: Run the 5-touch sequence
Templates below. Edit voice to match yours. Keep sentences short.
Step 4: Measure reply rate per touch and tune
After 50 to 100 runs, you will know which touch is doing the work. Usually touch 1 produces the bulk of replies. Touch 3 (the resource share) tends to produce the second wave. If touch 5 is producing nothing, shorten the sequence.
Step 5: Decide on automated vs rep-sent
For Tier 1, every touch should be rep-sent from the rep’s address. For Tier 2, touches 1 and 5 are rep-sent and touches 2, 3, 4 can be semi-automated with rep approval. For Tier 3, run the full sequence from a tool like Outreach or Salesloft with visitor data populated into custom fields.
The templates
Touch 1 - Day 0 - Email
Purpose: open with relevance, not pitch. Offer permission.
Subject: [Company Name] + [category]
Hi [First name],
Reaching out because I work with [3 similar companies] on
[specific problem in their category]. If [Company Name] is
anywhere near evaluating [category], I can share how they are
thinking about it.
Happy to send a 2-minute writeup or to jump on a 15-minute call,
or to just leave you alone if the timing is off.
- [Rep first name]Why this works: the signal that you reached out is implicitly “we see you are in the category.” The explicit frame is “here is a helpful peer comparison.” The out is baked in (“leave you alone if timing is off”). No one feels watched.
Touch 2 - Day 2 - LinkedIn connect, no pitch
Purpose: social proximity. Do not attach a pitch to the connection request.
Hi [First name], working in [category] space and saw your profile.
Happy to stay connected.That is the whole thing. If they accept, you have a second channel. If they do not, nothing changes.
Touch 3 - Day 5 - Email (resource share)
Purpose: prove value before asking for anything.
Subject: [Specific artifact] for [Company Name] - probably useful
Hi [First name],
Not a follow-up email.
[3 similar customers] built a [specific artifact: ROI model, comparison
grid, buying-committee brief] for [category]. I rewrote it into a
version that maps to [Company Name]'s size / industry.
Link: [direct link, no form wall]
If it saves you 20 minutes of evaluation time, great. If not, delete
and no follow-up needed.
- [Rep first name]Why this works: genuine value, no form fence, no meeting ask. The “not a follow-up email” line breaks the expected cadence. Reply rate on touch 3 tends to be the second-highest of the sequence.
Touch 4 - Day 8 - LinkedIn message (post-connect only)
Purpose: soft ask. Skip if they did not accept the connect.
Hi [First name],
Saw the [industry news / podcast / report] from last week and
thought of you. If you're still thinking about [category], the
takeaway on [specific point] lines up with what our customers
have been saying.
Happy to send a short writeup if useful.Why this works: references a public event in the world, not a private event on your site. Keeps the conversation in-category without creepy hooks.
Touch 5 - Day 11 - Email (direct ask, short)
Purpose: close the loop. Assume the close.
Subject: 15 min next week?
Hi [First name],
Three quick options:
1. 15-min call next Tues/Wed to walk through how
[3 peer companies] evaluated [category]
2. I send a one-pager, you read it when you have time
3. Not a priority right now, circle back in Q3
Which one?
- [Rep first name]Why this works: it’s a forced choice. Even “option 3” is a positive signal because it gives you a clean rotation back into the nurture at a named date. This touch tends to close the sequence cleanly, either as a booked meeting or as a “not now” with a date.
Sample Slack alert tied to sequence launch
:zap: Warm-outbound sequence launched
Who: Morgan Rios · Head of Growth · Ferry Labs (72 FTE)
Source: identified on /pricing, ICP match
Tier: Tier 1
Sequence: warm-outbound-v3
Rep owner: @tara
Stop if: reply, unsubscribe, OOO bounce
Next touch: Apr 26 (LinkedIn connect)Log every touch. If the prospect replies, stop the rest of the sequence immediately.
Common failure modes
Referencing the specific page. “Saw you visited our pricing page” is the fastest way to kill the conversation. Reference the category instead.
Treating it like a cold sequence. Using the same 9-touch cadence with the same cold-list template. They already know who you are, they need a different register.
Sending from a no-reply or marketing address. Warm outbound must look 1-to-1. Rep’s own address, no tracking pixel if possible, no auto-signature block full of marketing links.
Auto-personalization that goes wrong. “Hi {first_name}” with a visible bracket because the merge field broke. Always preview before sending.
Ignoring time-zone of recipient. Touch 1 at 2am local time reads as automated. Send during their business hours.
Skipping the suppression check. You will send this to a current customer. The customer will screenshot it. The screenshot will reach their CSM. Do not skip suppression.
Running this on EU/UK visitors without consent. Leadpipe defaults to company-level data for EU/UK visitors. Person-level outreach to EU contacts requires you verify consent upstream or rely on B2B legitimate interest with opt-out. Run the sequence against the company, not the individual, until consent is confirmed.
Not stopping on reply. The single most embarrassing error. Touch 3 lands after the prospect already replied to touch 1. Your sequence tool must stop immediately on any inbound reply.
Measurement: how to know it is working
Benchmarks for a well-run warm outbound sequence to identified visitors:
| Metric | Target |
|---|---|
| Open rate touch 1 | 55 to 70% |
| Reply rate touch 1 | 10 to 15% |
| Overall reply rate across sequence | 15 to 25% |
| Meetings booked per 100 entered | 6 to 12 |
| Opt-out / unsubscribe rate | Under 3% |
| Negative replies (“stop”, “who are you”) | Under 2% |
Compare that to a cold sequence to unidentified list contacts: 1 to 3% reply rate, 1 to 3 meetings per 100, 2 to 5% opt-out.
If your opt-out rate is above 3%, you are probably referencing visit data too explicitly. Audit the copy. The reference rule is the load-bearing part of this whole sequence.
If your reply rate is below 10%, the targeting is off, not the copy. Re-audit the ICP filters.
Where this fits in the bigger system
The warm outbound sequence is the action layer on top of identified visitors. Identification gives you the data, routing gives you the owner, the follow-up SLA gives you the clock, and this sequence gives you the touches.
Pair it with:
- Auto-routing so the right rep is running it
- Slack alerts for Tier 1 handoff
- Salesforce or HubSpot for sequence enrollment
- Outreach or Salesloft for execution
On US B2B traffic Leadpipe identifies 30-40%+ of visitors deterministically with full contact data. The independent accuracy test put us at 8.7/10 against RB2B at 5.2 and Warmly at 4.0. Quality of sequence output is capped by quality of identification input.
Leadpipe identifies 30-40%+ of your US B2B visitors with full contact data on the Pro plan at $147/mo. No credit card to start the 500-lead trial. Start identifying visitors →
