One of the most common questions from companies starting a cold email program is a simple one: when does it start working?

The honest answer is that the timeline has more structure to it than most people expect. There is a warmup period, a ramp period, and a compounding period. Knowing what to expect at each stage prevents teams from abandoning a working program too early — or from continuing with a broken one too long.

The setup period: weeks 1–2

Before a single prospecting email goes out, there is infrastructure work that must happen. This period is not glamorous and produces no pipeline, but it determines whether the subsequent campaign lands in inboxes or spam folders.

What happens in weeks 1–2:

• Dedicated sending domains are registered and DNS authentication records (SPF, DKIM, DMARC) are configured.
• Sending inboxes are created and the warmup process begins.
• The target list is built, verified, and segmented.
• The email sequence is written and reviewed.

Nothing in this period is visible to prospects. It is all foundation.

Inbox warmup: weeks 2–4

Warmup takes 2–4 weeks depending on the tool and target volume. During this period, the sending inboxes are gradually building a positive sending history, with send volume increasing incrementally each day.

Warmup cannot be compressed without risking the deliverability of your campaigns. A domain that has not been properly warmed up will see open rates of 5–10% and near-zero replies — not because the copy is bad, but because the emails are landing in spam.

At the end of warmup, run a seed test to verify inbox placement is acceptable across Gmail, Outlook, and other major providers before proceeding.

Month 1: the first real sends

By the time you reach month one of actual outreach, you are 3–5 weeks into the process. This is when the first campaign launches.

What to expect:

• Open rates should be in the 40–60% range if warmup was done properly.
• Replies will start coming in within the first few days. Do not expect the bulk of replies immediately — most come from follow-up emails sent in weeks 2 and 3 of the sequence.
• Meetings may start booking, but month one is typically light on conversions. You are gathering data more than you are building pipeline.
• We will identify problems: addresses that bounced more than expected, subject lines that underperformed, a segment that is not responding.

The right posture in month one is not impatience. It is curiosity. Every data point tells you something.

Month 2: first signals of what works

By month two, you have one complete campaign behind you and you are launching the second. The pattern becomes clearer:

• Which segments responded and which did not.
• Which email in the sequence drove the most replies.
• Whether the positive reply rate is directionally strong or weak.
• Whether the offer resonates with the specific role and company type you targeted.

Month two is typically when the first meetings that originated in month one start converting. Pipeline from month one's outreach matures through discovery calls and follow-up conversations.

We also iterate on the copy based on month one data. If email two in the sequence had a dramatically lower open rate, the subject line might be the culprit. If you got replies from a specific subsegment but not others, the next campaign should lean into that subsegment.

Month 2 benchmark: If you have sent 400–600 emails across two campaigns and have not booked a single positive conversation, that is a signal that something is structurally wrong — either with the list, the infrastructure, or the offer. It warrants a deeper diagnostic before continuing.

Month 3: compounding returns

Month three is where consistent programs start to look like an actual engine. You now have:

• Two or three complete campaigns with performance data
• An informed ICP based on who actually responds
• Refined copy based on what worked and what did not
• A warmup infrastructure that is running smoothly
• Meetings from previous campaigns converting into opportunities

This is also the point where teams that have been running well start to see compounding effects. A prospect who ignored your campaign in month one receives a new campaign in month three with a different angle — and this time the timing is right. Pipeline attribution starts to feel less clean because touches from different campaigns are influencing the same prospect.

Month 3 benchmark: A well-running program at this stage should be generating enough meetings per month to represent a meaningful contribution to your pipeline — typically 8–20 meetings per month depending on list size, offer, and market, for an active program sending 1,000–2,000 emails per month.

What can extend the timeline

A few factors extend the timeline beyond what is described above:

ICP is not yet defined. If you are still figuring out who your ideal customer is, the first two months will teach you more than they produce in pipeline. Build for learning, not just output.

The offer needs repositioning. Sometimes the issue is not the email — it is that the value proposition does not resonate with the market you are targeting. This takes longer to identify and fix than a deliverability problem.

Inconsistent execution. Cold email programs that run sporadically — a campaign here, two weeks off, another campaign, another gap — do not compound. Consistency matters. A program that sends three campaigns per month for six months will dramatically outperform one that sends one campaign per month over eighteen months.

The question to ask at 90 days

After 90 days, the right question is not "is cold email working?" That question is too broad to be useful.

The better questions are:

• Have we confirmed the infrastructure delivers to the primary inbox?
• Have we identified at least one segment that responds to our outreach?
• Are the conversations we are booking converting into qualified opportunities?

If you can answer yes to all three, the program is working and the job is to scale what is working. If you cannot answer yes to any of them, something structural needs to change before adding more volume.

The timeline is predictable. The results are not automatic.