Michel Lieben runs ColdIQ, a $6.5M ARR outbound agency, and in mid-August 2026 he published a case study that's worth more than most of the multichannel cold outreach advice circulating right now, because it isn't a theory. It's a before-and-after on a real client campaign, with a number attached: reply rate went from 1.8% to 4.3%. Nothing about the offer changed. Nothing about the copy changed. One thing changed: the order the channels fired in.
Inside the ColdIQ multichannel cold outreach case study
Most teams already run more than one channel. They run LinkedIn and email at the same time, on the theory that more touchpoints in more places equals more attention. Lieben's case study argues that's the wrong way to think about a multichannel cold outreach program. Channel count isn't the lever. Channel order is. The campaign in question started at a 1.8% reply rate running email and LinkedIn as separate, parallel tracks. After restructuring the sequence so LinkedIn ran first and email followed, the same campaign hit 4.3%.
The mechanic itself is simple enough to describe in one sentence: view the prospect's LinkedIn profile, send a connection invite, wait, then send the first cold email. No new list. No new offer. No new subject line testing. Lieben published the full breakdown on ColdIQ's blog, and the headline claim is specific enough to be checkable: this was a sequencing change on an existing campaign, not a new campaign built from scratch. That's what makes it worth a teardown instead of a footnote.
The old sequence: LinkedIn and email running in parallel
Before the change, the campaign treated LinkedIn and email the way most teams still treat them: as two lanes running at the same time, coordinated loosely if at all. A prospect might get a connection request on Monday and a cold email on Monday too, sent by a different tool, tracked in a different dashboard, with no logic tying the two together beyond both existing in the same go-to-market plan. Whichever channel happened to fire first was essentially random, decided by queue timing rather than by design.
That's not a lazy setup. It's the default setup, because most sequencing tools are built channel-first, not sequence-first. You configure an email cadence in one place and a LinkedIn cadence in another, and the two rarely share a trigger. The result is two channels doing the same job, cold contact, at the same time, without either one setting up the other. That's the 1.8% baseline: a real multichannel program, correctly staffed, that still treated multichannel as parallel rather than sequential.
The new sequence: LinkedIn first, email second
The fix wasn't a new channel. It was a new order, with a gap built in on purpose. LinkedIn goes first: view the profile, send the connection invite. Then a wait. Then the cold email lands, as the first message in the recipient's inbox from this sender, but not the first contact from this company. By the time the email arrives, the prospect has already had one small, low-friction interaction with a name attached to a face.
Reply rate before and after sequencing LinkedIn ahead of email (ColdIQ client campaign, Aug 2026)
The reasoning holds up without needing anything exotic. A cold email from a complete stranger asks the reader to trust a name they've never seen, in the two seconds before they decide to delete it. A cold email that arrives after a LinkedIn profile view and a connection invite asks less of the reader, because the name isn't new anymore. It's a small prior. The recipient doesn't have to consciously remember the LinkedIn touch for it to work; passive familiarity is enough to change how the next message gets read. That's a sequencing effect, not a persuasion trick built into the copy.
What changed operationally to make multichannel outbound reply rate gains stick
The idea, LinkedIn before email, is not new. Plenty of outbound playbooks have suggested it for years. What's different in this case study is that it got implemented as a forced sequence with a trigger, not as a loose guideline reps were supposed to remember. That's the part worth copying, and it's mostly a workflow change, not a strategy change.
| OPERATIONAL ELEMENT | OLD APPROACH | NEW APPROACH |
|---|---|---|
| Trigger for email send | Fires on its own schedule, independent of LinkedIn status | Fires only after the LinkedIn view + invite step completes, plus a gap |
| Who owns the LinkedIn touch | Loosely assigned, sometimes skipped under time pressure | Built into the sequence as a required, non-skippable step |
| Sequencing logic | Two parallel cadences with no shared dependency | One sequence, single source of truth, channel order enforced |
| Gap between touches | Not defined; both channels could land same day | Defined wait between LinkedIn touch and email send |
| Reporting | Reply rate tracked per channel, not per sequence | Reply rate tracked against the full sequence, including channel order |
Notice what's absent from that table: nothing about better subject lines, nothing about a new personalization angle, nothing about list quality. The entire lift traces back to enforcing order and building in a gap. That's a deliberately narrow claim, and it's also why it's believable. A vague claim like "we improved multichannel cold outreach" is easy to dismiss. A specific, mechanical change, invite-then-wait-then-email, enforced by sequencing logic instead of by hoping a rep remembers, is the kind of thing you can actually go implement this week.
This is the same discipline we push on inside any cold email program we run: the channel mix matters less than most teams assume, and the coordination between channels matters more. We've made a version of this argument before when breaking down how channel mix actually functions inside an outbound funnel, and this case study is a clean, isolated example of the same principle: adding LinkedIn as a parallel channel is not the same intervention as sequencing it in front of email on purpose.
The infrastructure a LinkedIn plus email sequence actually requires
None of this works if the underlying infrastructure can't support it. Sequencing LinkedIn before email at any real volume runs into limits that a single-rep pilot won't surface. Here's what has to be true operationally before a team can run this beyond a handful of accounts.
Deliverability sits underneath this too, even though it isn't the headline of Lieben's case study. A cold email that lands after a LinkedIn touch still has to land in a primary inbox to get read. If your sending domain, warmup, and volume controls are already shaky, layering a LinkedIn step in front of the email won't rescue a message that's landing in spam. We've laid out the baseline requirements for that separately in our cold email deliverability playbook, and it's worth checking that box before adding a new sequencing layer on top of it.
The honest caveat, and how to test it before you roll it out
One case study is one case study. A 2.4x reply-rate lift on a single client campaign is a strong result, and it's also a result sensitive to variables Lieben's post doesn't fully control for: list quality, the specific offer, the starting baseline, the industry, and the size of the list itself. A campaign starting at 1.8% has more room to move than one already starting at 4% or 5%. None of that makes the finding worthless. It makes it a hypothesis worth testing on your own list, not a rule to roll out to every active sequence tomorrow.
“The lift is real. The multiplier is not a guarantee. Test the sequencing change on a slice before you touch every live campaign.”
There's also a scale problem worth naming plainly. LinkedIn seat limits mean this sequence doesn't scale the same way pure email does. A team that can send thousands of cold emails a week from a handful of warmed domains cannot send thousands of connection invites a week from a handful of seats without tripping restrictions. Any team adopting this has to accept a lower total volume ceiling, or add more seats, both of which change the cost math of the campaign, not just its reply rate.
It's also worth putting this case study next to the broader argument about whether outbound volume itself is the problem, an argument we've dug into in our breakdown of the AI-volume outbound debate and in another teardown of an inbound-to-outbound sequence. The throughline across all of it is the same: the lever that moves reply rate is rarely more volume or a cleverer line. It's usually structure, who gets contacted, in what order, with what signal already in place before the ask arrives. Sequencing LinkedIn ahead of email is one version of that structural fix. It happens to be a well-documented, specific one, with a real before-and-after number attached to it.
Do this next: pick one live campaign, carve out a slice, and rebuild the sequence so the LinkedIn view and connection invite fire first, with a real gap, before the email goes out. Keep everything else identical. Give it two to four weeks, check the reply rate against a control segment running your current parallel setup, and only expand it past the pilot once you've confirmed the lift holds on your own list, not just on ColdIQ's.
See where you are cited today
A free snapshot audit of your rankings and AI citations before we ever talk.
Tyler leads work at the intersection of SEO and generative engines at Something Inc., helping B2B brands get ranked and cited across every major AI engine.