Running an ABM Play End to End
Outcome: a running weekly motion where each tier gets a different treatment, signals decide the timing, and the whole thing reports on itself.
- Surface
- App and MCP server
- Level
- Intermediate
- Uses
- Everything from lessons 02–07
- Credits
- ~40 per week for a 500-account list
- Prerequisite
- Lessons 06 and 07
Different tier, different motion
| Tier 1 | Tier 2 | Tier 3 | |
|---|---|---|---|
| Accounts per rep | 20–50 | 100–300 | The rest |
| Research | Individual, deep | Segment-level | Automated only |
| Contacts | 4–6, multi-threaded | 2–3 | 1–2 |
| Message | Written per account | Segment template + evidence line | Fully generated |
| Channels | Email, phone, LinkedIn, events, exec-to-exec | Email, LinkedIn | |
| Signal role | Decides when to act | Decides whether to act | Decides whether to act |
| Cadence | Continuous | Signal-triggered | Signal-triggered |
The distinction that matters is in the last two rows. Tier 1 accounts are worked whether or not a signal fires — the signal only changes the timing and the angle. Tier 2 and 3 accounts are worked only when a signal fires.
Multi-threading Tier 1
A single contact at a strategic account is a single point of failure: they leave, they go quiet, they were never the decision maker.
Map the committee
Four to six people across the roles from TAM Sourcing lesson 06: practitioner, champion, economic buyer, technical evaluator.
Different message per role
The practitioner cares about the work; the buyer cares about the outcome. The same email to both is wrong for at least one.
Sequence the entry
Not all at once. Champion first, then widen based on response — or lack of it.
Track the account, not the contact
Reply from anyone counts as account engagement, and should pause competing outreach to the others.
Multi-threading without coordination reads as a mass email. Four people at one company receiving near-identical messages in the same week will notice — and they will forward them to each other. Different angle per role, staggered timing, and a shared record of who has been contacted.
The weekly rhythm
Monday — signals run
All signals across the target list. Scores recompute with decay applied.
Monday — the queue is produced
Accounts that newly crossed the threshold, with the signal that pushed them over.
Tuesday — Tier 1 review
A human reads the Tier 1 movers and decides the angle. Twenty minutes, not a meeting.
Tuesday — Tier 2 and 3 dispatch
Automated. Evidence-driven copy, verified contacts, gated on QC.
Ongoing — replies routed
To the account owner, with the triggering signal attached.
Friday — the numbers
Signals fired, accounts triggered, contacted, replied. Five lines.
The rhythm is what makes it a motion rather than a project. Signals fire whether or not anyone is watching; the rhythm is what turns them into action.
What to measure
| Metric | What it tells you |
|---|---|
| Signals fired per week | Is the pipeline alive? |
| Accounts crossing threshold | Is the scoring calibrated? |
| Contact rate on triggered accounts | Is enrichment keeping up? |
| Reply rate by signal type | Which signals are worth keeping |
| Reply rate by tier | Is the extra Tier 1 effort paying? |
| Time from signal to first touch | The number that decides whether timing advantage is real |
The fourth row is the one that improves the program. After a quarter you will find that one or two signals produce most of the replies and one produces almost none — and dropping the latter makes everything cheaper without costing anything.
Do this now
Assign every account a tier
Write the treatment per tier
What research, how many contacts, which channels, which message type.
Map the committee for Tier 1
Four to six people, per role.
Schedule the signal run
Weekly, all signals, decay applied.
Build the queue view
Newly crossed accounts, with the triggering signal.
Set the human review slot
Tier 1 only, twenty minutes.
Automate Tier 2 and 3 dispatch
Gated on QC and suppression.
Build the Friday summary
Six numbers, delivered to Slack.
Check your work
- Every account has a tier and a defined treatment
- Tier 1 accounts are multi-threaded with role-specific messages
- The weekly rhythm is scheduled, not aspirational
- Reply rate is tracked by signal type
- Time from signal to first touch is measured
Where this breaks
Treating every tier the same defeats the purpose in both directions: Tier 1 accounts get automated messages that waste the commitment you made to them, and Tier 3 accounts get research effort they cannot repay. If the only real difference between your tiers is a label in the CRM, you are running outbound with extra reporting — which is fine, but it should be a decision rather than a drift.
Course complete
You now have a signal catalogue with windows and cadences, three complete trigger plays, a tiered target account list with entry and exit rules, a decayed cross-signal score, and a weekly operating rhythm that reports on itself.
Where to go next:
| You want | Course |
|---|---|
| Hiring as a full course | Hiring Signals |
| Champion tracking as a full course | Job Changes & Champions |
| The account universe underneath | TAM Sourcing |
| To run the plays from chat | AI for Sales Reps |
Reference for this lesson: Signals, CRM integrations, Outreach integrations, Workspace integrations.