The first 90 days of cold outreach for a 20 person integrator

We get asked some version of this every week. We are a twenty person integrator, we have always grown through relationships and referrals, and the owner just said it is time to take cold outbound seriously. Where do we start. The honest answer is that the first ninety days are not about leads at all. They are about building the smallest possible machine and learning to operate it without flinching.

Days 1 to 14, get the plumbing right

Before you send a single email, set up the boring layer. You need at least three sending domains, all distinct from your main brand domain. You need mailboxes warmed for two to three weeks before any real volume. You need a single source of truth for every contact you touch, so that two reps do not email the same person within the same quarter.

• Buy three sending domains in the .com or .net space, with names that read like a person, not a marketing campaign
• Set up SPF, DKIM, and DMARC the right way the first time. Half-finished DNS is the most common reason new teams get spam-foldered
• Pick a unified inbox so replies do not get lost across the three domains
• Decide who owns suppression. One person, one source of truth, no exceptions

Days 15 to 45, define what good looks like

If you spend the first month sending, you will spend the second month confused about results. Spend the first month writing down the smallest possible definition of an ideal account. Two pages, not twenty. Include the verticals you actually want to win and the verticals you tolerate. Include the deal size you are willing to do and the deal size below which you are losing money on every truck roll.

Then write three sequences. One for permits. One for leadership changes. One for expansions or new builds. Do not write a generic introduction sequence. The generic sequence is what every other integrator is sending. Specificity is your only defensible weapon when you are 20 people up against a 200 person regional.

Days 46 to 75, send small and read everything

First real sends should be small. Twenty to forty per domain per day, ramping by ten percent weekly. Read every single reply for the first month, including the ones that say no. The negative replies will teach you more in three weeks than any course on copy will teach you in a year.

Days 76 to 90, calibrate, then commit

By day seventy five you should have enough data to look at three numbers honestly. Reply rate per sequence. Meeting set rate per reply. And the cycle time from meeting set to a real proposal. If reply rate is below 4 percent, your targeting is wrong. If meeting rate from reply is below 30 percent, your handoff is bad. If cycle time is over six weeks for a deal under 100k, you are chasing accounts that should not have been chased.

Most integrators that fail in their first quarter fail because they tried to scale before they had read a single batch of replies carefully. The teams that succeed look almost identical at day ninety. Small numbers, clean data, one rep who knows the territory, and a stack of negative replies they have already learned from.