Agencies that succeed with outbound rarely stumble into it. They follow a sequence — foundation, then launch, then scale — and resist the urge to skip straight to sending because sending feels like progress. Here's the 90-day version of that sequence.
Get the unglamorous parts right first
Define the ICP precisely enough that you could name five real companies that fit it. Set up sending infrastructure on a dedicated subdomain and begin domain warm-up — see deliverability basics if you haven't yet. Draft messaging around the specific problem your ICP has, not a generic value proposition. Build (or verify) an initial list of 300–500 tightly-matched prospects.
Send in small batches, measure, adjust
Launch to a small daily volume — enough to gather signal, not so much that a bad early message burns the whole list. Track reply rate and, more importantly, positive reply rate and booked-call rate, not opens. Run at least two message variants per ICP segment and kill the underperformer fast. This is also the phase to add a second channel — see the multi-channel playbook — once email alone has a baseline.
Increase volume only after the system proves itself
Once messaging and targeting are validated at small scale, increase sending volume and expand the prospect list to adjacent ICP segments. Build the intent-signal layer — see how to know when to reach out — so outreach starts triggering off real buying signals instead of a flat cadence. By day 90 the goal isn't a single successful campaign; it's a repeatable system that keeps producing qualified calls without needing to be rebuilt from scratch every quarter.
The agencies that get outbound wrong almost always skipped Phase 1 to get to Phase 2 faster. The ones that get it right treat the first 30 days as infrastructure, not delay.
Why 90 days, specifically
Shorter timelines don't give deliverability warm-up or message iteration enough room to work — you end up scaling a system you haven't actually validated. Longer timelines tend to lose organizational momentum before the system produces enough evidence to justify continued investment. Ninety days is long enough to build something real and short enough to stay accountable to results.
See what this looks like as a real case example in our web agency case study, or book a call and we'll map this framework to your specific ICP.