The Order I'd Launch 5 Outbound Signal Plays In (The Best One Isn't First)

I build outbound systems for B2B clients, and the order you turn on five different signal plays changes the outcome almost as much as which five you pick. Most breakdowns of this topic stop at the plays themselves. The part that's actually useful, and the part a video walkthrough glosses over because it's shot as one continuous list, is which one goes live on week one and which one waits until the rest is already running on its own.

Why cold lists are dying and signals aren't

The old way of running outbound was to buy a list of a few thousand contacts, send the same email to all of them, and accept a reply rate somewhere between 1% and 3%. That still works in the sense that it still sends, but it burns domains fast and the majority of the list never had a reason to care in the first place.

A signal flips the order of operations. Instead of starting from a title or an industry, you start from something a company just did: they closed a funding round, hired into a specific function, visited a pricing page, or started following a competitor. The message gets built around that action, the list is smaller, and the reply rate moves into double digits because the person reading it has an actual reason to recognize the email as relevant rather than generic.

The five plays, side by side

Five plays, each reacting to a different kind of signal. The reply rate figures below are the ranges given in the video itself, not a guarantee, and I've flagged the one that isn't a flat number.

Play 1: lookalike campaigns

Take your 10 to 50 best closed-won accounts, work out what they have in common, stage, size, a recent hire, a specific trigger, and build a list of companies that match the same pattern. A tool such as Ocean.io can pull the firmographic matches, Clay does the enrichment and scoring, and the pitch writes itself: you worked with a similar company and got a specific result, would they want to know how. Because the comparison is real, the reply rate holds at 8 to 12% even on cold email. LinkedIn's own lookalike audience feature is gone, so most of this now happens through third-party matching tools instead; I went through the current options in more detail in lookalike targeting for B2B clients.

Play 2: website visitor tracking

A pixel-based tool like Albacross can identify a visitor down to the company and, in some cases, the individual. This one works best for a US-heavy pipeline. European privacy rules make identifying an individual visitor by name a lot harder to do safely, so if most of your traffic is from the EU, this play contributes less than the other four.

The part worth taking seriously: track four or five strategic pages, pricing, case studies, maybe a comparison page, not every page on the site. Route the matches through Clay for ICP scoring, push qualified visitors to Slack and a sequencer, and reach out the same day, whether or not the message actually mentions the visit. Albacross isn't the only option here either; I compared two other identification tools, including one that goes after the same job in a different way, in RB2B vs Common Room for website visitor identification.

Play 3: social listening

This one monitors LinkedIn, X and Reddit for a company complaining about a competitor, posting about a pain point, or going through a hiring spree, and the goal is to reach out within 24 hours of the post going live. The tool I'd have pointed you to for keyword monitoring a few months ago, the one named in the video, is shutting down. Its team joined HubSpot and the standalone product closes on 22 October 2026, which is five days from when I'm writing this. I went through what it did and what replaces each of its jobs in its shutdown and the alternatives, job by job, so check that before you build a new social listening setup around a tool that's about to stop collecting signals.

The underlying play still holds regardless of which keyword tool runs it: identify 5 to 10 trigger phrases, route qualified posters to a connection request, and send the message the moment they accept so there's no delay between the two steps. Some clients have seen this play land a 50 to 60% reply rate, which is the highest of the five, and also the reason it isn't the one to launch first.

Play 4: competitor and complementary tool followers

Identify your top competitors, or tools that solve an adjacent problem for the same buyer, and pull their LinkedIn followers with a scraping tool such as Leadsforge, filtered down to the job titles and geography that actually match your ICP. Pulling every follower of a 100,000-follower page wastes credits and time; filtering first is what makes the list usable.

Against a competitor, the opener can be direct: ask if they're currently using that tool for the problem you solve. Against a complementary tool, the angle flips from competing to partnering: you integrate with it, or add a capability it doesn't have. Both versions get a 10%+ reply rate when the filtering narrows the list to real fits, and the video's own claim is that 10% isn't even the ceiling if the copy is right.

Play 5: event outbound

Find the people attending an industry event, either from an official attendee list or by tracking who's posting about going, and run three campaigns timed to the event: before, during and after. A pre-event message as simple as naming the event and suggesting a coffee gets an outsized number of yes replies, because people go to events specifically to make new connections and they're primed to say yes to exactly that ask. Around a 50% reply rate is realistic here, and it climbs higher the narrower and more relevant the event is to your actual buyer.

The order I'd actually launch these in

Rank the five plays by reply rate and social listening and event outbound come out on top. That's not the order I'd launch them in, and it isn't the order the video recommends either. The sequencing logic is about what each play costs to stand up and maintain, not what it returns once it's live.

Lookalike campaigns go first because the data already exists in your CRM. No new tool, no new signal source, just a list build from accounts you've already closed. Website visitor tracking goes second specifically because it needs 2 to 3 months before the page-level pattern is reliable, so the sooner that clock starts, the sooner it compounds. Social listening goes third: best reply rate on the list, but it needs daily attention on the keywords, and it's wasted effort to hand someone that workload before the first two plays are already running without supervision. Event outbound slots in fourth, triggered by the calendar rather than a weekly cadence, so it starts whenever the next relevant event is roughly two weeks out. Competitor and complementary tool followers come last, scaled only once the first four are already running, because it needs the most iteration on the copy to work well.

If you want a structure for testing each play properly before you stack the next one on top, rather than launching all five in the same week and not knowing which one is doing the work, I laid out a five-stage process for that in how to run outbound experiments without wasting a month.

Why two of these plays skip email entirely

In my own sequences, email and LinkedIn almost always run as one motion rather than two separate efforts aimed at the same person. Event outbound and, to a lesser extent, social listening are the exception worth calling out. For event outbound, a LinkedIn connection request and message is often the whole sequence, no email leg needed, because the volume is small and the ask (meet at the event) works better as a personal message than a cold email. Social listening leans the same way when the list that week is a handful of people, not hundreds.

The other three plays (lookalike, website visitors, competitor followers) run at a volume where email earns its place back in the sequence. The rule isn't "use every channel on every play." It's "match the channel to how many people are actually on the list this week."

Three patterns that hold across all five

• Relevance over volume. None of these five plays need a large send volume to produce results. Smaller, signal-driven lists consistently outperform a bigger blast, which also means your sending infrastructure doesn't need to carry much.
• Timing beats copy. A fresh signal, acted on within a day or two, outperforms a polished message sent to a stale list. If a campaign underperforms, check how old the signal is before you rewrite the email.
• They stack. Combine social listening with event outbound, or competitor followers with a lookalike list, and the more signals pointing at the same account, the better the message and the higher the reply rate tends to land.

What this looks like over your first 90 days

Weeks 1 and 2: pull the closed-won list, build the lookalike segment, and get the first lookalike campaign live. Weeks 3 and 4: install the website visitor pixel, pick four or five pages to track, and let it run in the background while you move on. By week 5 or 6, start social listening: pick your keywords, connect the alert to Slack, and commit to checking it daily since this is the play that needs ongoing attention. From week 7 onward, watch your industry's event calendar and launch an event campaign whenever something relevant is roughly two weeks away, pre-event, during, and a follow-up after. Competitor and complementary tool followers come in around week 8 or later, once the first four aren't taking active daily management anymore and you have the time to get the copy right.

None of this needs to happen in one sprint. The order exists so you're not running five half-built systems at once with nobody checking any of them properly.

FAQ