# 16 GTM Plays With Receipts

Most campaign libraries are lists of ideas. This one only contains campaigns somebody has published a result for, with a link to where they published it. If a play is not in here, it is usually because nobody has written down what it did.

- Every campaign has a trigger, a build, and one number. If a play has no trigger, it is a list, not a campaign.
- Read the number next to who published it. Most GTM benchmarks are vendor-published, and the vendor always wins their own benchmark.
- Your own cold baseline beats any published average. Hold out a control group or you cannot claim anything.
- Build one campaign properly before you build five badly. Depth in one channel is what gets hired.

## Play 1. Tier your own connections every month

Stack: FullEnrich, lemlist, Attio

Your network is already a list, it is just an unsorted one. Once a month the agent pulls every LinkedIn connection, sorts them into three tiers, enriches the ones worth contacting and writes them into the CRM. Then it asks the only question that matters: is any of these a live opportunity nobody is working? Most people treat their connections as a vanity number. It is a list you already earned the right to email.

## Play 2. Test one hypothesis a month on 100 accounts

Stack: Yalc.ai, Slack, lemlist

Pick 100 accounts. Derive five to ten signals that would mean something if they fired: they are hiring, they run a certain tech, they just cut headcount, a competitor of theirs folded. Wire an agent to watch all of them and ping Slack the moment one fires. The point is not the 100 accounts. The point is that a month later you know which signals were real and which you imagined, which is the only way a trigger portfolio ever gets good.

## Play 3. Rebuild the prospect list from what you actually closed

Stack: Attio, Yalc.ai, lemlist

Every month the agent reads what closed in the CRM, finds companies that look like those, writes them back as prospects and drafts the campaign. You verify before anything sends. This is the cheapest list in the business because it is derived from evidence rather than from a filter somebody guessed at. The companies that bought from you are the best description of your ICP you will ever write.

## Play 4. Find the gap between your story and what buyers say

Stack: Claap, Yalc.ai

Once a month an agent reads every call you had with prospects and holds it against your outreach copy, your campaigns and your written ICP. Anywhere the two disagree is something to fix before it costs a deal: the wrong qualification criteria, a promise the website makes that the call cannot keep, a pain your buyers name that your copy never mentions. Nobody runs this play, which is exactly why it works.

## Play 5. Turn attention on your content into a list

Stack: Yalc.ai, lemlist

People engage with a post, visit your profile, follow you or the company page. Each of those is a small, cheap intent signal that almost everybody throws away. Capture all of them and route them into a value-first campaign: a guide, an asset, a webinar you already recorded. You are not pitching anyone. You are handing something to a person who just raised their hand, which is the entire reason this page you are reading exists.

## Play 6. Steal the campaigns your competitors already published

Stack: Yalc.ai

An agent reads the blogs of the biggest lead-gen agencies and products, pulls out every campaign they have described launching, and diffs that against the list of things you have actually tried. What comes back is a queue of unscalable, unsloppy plays nobody on your team would have thought of. They published it, which means they already paid for the lesson. You just have to run it.

## Ten more, taken from published teardowns

## 1. Champion job change (Outbound)

**Fires when:** Someone who used your product at a previous company starts a new role.

1. Sync every closed-won and power-user contact into a watch list.
2. Poll for employer changes weekly, not daily. Nobody buys in week one.
3. Gate on two things: the new company is in ICP, and the person kept or gained seniority.
4. Open with what they already know about the product, never with congratulations.

**Result:** 17% or higher reply rate and a 10% booked-meeting rate, roughly 2x Common Room's cold baseline. Former champions convert about 3x more often than cold prospects.

Source: [Common Room, Plays that pay](https://www.commonroom.io/blog/track-job-changes-to-fuel-pipeline-growth/) (first-party)

## 2. De-anonymised website visitor (Inbound-led outbound)

**Fires when:** A known-account visitor hits a high-intent page and never fills in a form.

1. Accept that person-level resolution tops out around 15 to 20% of traffic. Work the account level for the rest.
2. Score by page, not by visit count. Pricing and docs beat the blog.
3. Because you cannot be certain who visited, lead with account research rather than the visit itself.
4. Route to a human within the hour. The window is the whole advantage.

**Result:** 2x reply rate and 2x booked meetings versus cold outbound, at 17%+ reply and 10% meetings. For context, about 98% of site visitors leave without filling anything in.

Source: [Common Room · HubSpot](https://www.commonroom.io/blog/deanonymize-website-activity-to-book-more-meetings/) (first-party)

## 3. The 48-hour trigger window (Outbound)

**Fires when:** Any buying signal fires: funding, a new exec, a tool change, a job post.

1. Treat latency as the product. Measure hours from signal to first touch.
2. Hold the record in a delay queue if the signal is noisy on day one, then send inside the window.
3. Never mention the signal itself. Say the consequence.
4. Instrument signal-to-meeting rate per trigger type so you can kill the dead ones.

**Result:** Signal-triggered campaigns report 3x to 5x higher positive response than static list outbound, and reaching a prospect within 48 hours of the trigger lifts booked meetings by up to 40%.

Source: [Vanderbuild, The signal revolution](https://www.vanderbuild.co/blog/the-signal-revolution-how-to-implement-b2b-intent-signals-in-outbound-campaigns)

## 4. Prune the trigger portfolio (Operating cadence)

**Fires when:** Quarterly review of every trigger you run.

1. Start with three to five trigger types, not fifteen.
2. Run each for a full quarter before judging it.
3. Keep only triggers whose signal-to-meeting rate clears 3%.
4. Expect a wide spread: funding and headcount growth carry most programmes, social engagement carries almost none.

**Result:** Reported conversion by trigger: funding around 30%, headcount growth 25%, a competitor post about 5%. One team's six-month numbers were 4.57x ROI, 85+ SQLs, 6 closed deals and a 35% CAC reduction.

Source: [Reachly, Signal-based outbound playbook](https://www.reachly.co/blogs/signal-based-outbound-playbook)

## 5. Waterfall enrichment before send (Data)

**Fires when:** Any list leaves research and enters sequencing.

1. Never run a single provider. Coverage caps out around half your list.
2. Chain providers cheapest-first and stop at the first verified hit.
3. Verify before send, not after bounce.
4. Track cost per verified contact, not cost per credit.

**Result:** Single-provider email coverage runs around 50%. A waterfall takes the same list past 80% coverage, across 50+ Clay deployments.

Source: [GTMinds, Clay outbound playbook](https://www.gtminds.io/blog/clay-outbound-playbook) (first-party)

## 6. Pain-matched micro-segments (Outbound)

**Fires when:** A segment shares one specific, nameable pain.

1. Segment by the pain, not the firmographic.
2. Write one message per pain and refuse to reuse it across segments.
3. Keep volume high enough to read the result, low enough that every message stays unique.

**Result:** A 1,478-lead campaign built this way returned a 24.2% positive reply rate. A second, aimed at founders and CEOs across 1,798 leads, returned 30% positive replies.

Source: [Outbound Republic case studies](https://outboundrepublic.com/case-studies/) (first-party)

## 7. Stop reporting open rate (Measurement)

**Fires when:** You are about to put open rate on a dashboard.

1. Apple Mail Privacy Protection and corporate scanners pre-load tracking pixels, so opens are partly machine noise.
2. Report reply, meeting and pipeline instead.
3. If you must keep opens, halve them mentally before drawing a conclusion.

**Result:** Reported open rates average about 43%, but real human opens sit nearer 25 to 30%. Only 15% of marketers still treat open rate as a primary metric.

Source: [Salesmotion, Signal-based outbound metrics](https://salesmotion.io/blog/signal-based-outbound-metrics)

## 8. Know your real baseline (Measurement)

**Fires when:** Before you claim any campaign worked.

1. Hold out a control group from the same list, every time.
2. Compare against your own cold baseline, not a vendor's published average.
3. Record the hypothesis before the build, so the result can actually falsify it.

**Result:** Published B2B cold-email baselines cluster between 0.5% and 3.4% reply, and broad sequences to scraped lists sit under 1%. Only 3 to 5% of any addressable market is in an active buying cycle at a given moment.

Source: [Growleads, citing Woodpecker and Forrester](https://growleads.io/blog/signal-based-outbound/)

## 9. Human checkpoint on an automated engine (Operating cadence)

**Fires when:** Any campaign about to leave the building.

1. Automate research, enrichment, drafting and routing.
2. Keep exactly one human gate: a person reads the batch before it sends.
3. The gate is a sample, not every row, or it stops scaling.

**Result:** One agency runs about 90% of its outbound workflow on automation and still passes every campaign through a human checkpoint before send.

Source: [The Playbook Agency, The Clay stack](https://www.theplaybook.agency/post/clay-automation-stack) (first-party)

## 10. Signal programme versus cold control (Outbound)

**Fires when:** You are deciding whether signal-based outbound is worth the build.

1. Run the signal programme and a cold control on the same ICP, same quarter.
2. Measure reply and meeting rate on both.
3. Only then decide what to keep.

**Result:** Teams running signal-based outbound report 15 to 25% reply rates against roughly 3% for traditional cold email on the same kind of list.

Source: [Reachly, citing Unify](https://www.reachly.co/blogs/signal-based-outbound-playbook)

## Before you build any of these

Pick one. Build it end to end, including the measurement. Write down the hypothesis before you start so the result can prove you wrong. Then take the number you got, not the number on this page, into your next conversation. A candidate who can show what their system changed is rarer than one who can show what they built.

Nine of the ten sources below are published by vendors or agencies describing their own results. Treat the numbers as directional, not audited. Where a source reports its own first-party data rather than a survey, it is marked first-party.