GTM experiments for B2B SaaS teams

The cold outreach guide: validate what works before you scale

Most cold email advice starts with volume. This guide starts with a question: does this segment respond to this offer? Answer it with a few hundred emails you can trust, then scale only what works.

Written for B2B SaaS founders and GTM teams. Every chapter treats a campaign as a GTM experiment: prove the emails arrive, test a precise segment with one offer, read a verdict you decided on before sending, and only then turn it into a fully automated campaign.

Start reading Plan an experiment — free

Why validate before you scale

Sending more emails doesn't make an answer clearer. It makes a wrong answer more expensive.

A zero can be a DNS record

We sent 337 emails across 13 campaigns and got no replies. The market wasn't the problem: the sending IP had no reverse DNS record, so the emails weren't arriving. Check delivery before you judge the market.

Big lists hide the answer

A broad list mixes companies with different problems. The overall result describes none of them. A narrow segment with one offer tells you something you can act on.

Scaling a guess costs months

Add mailboxes and volume before you know an offer works, and you spend a quarter learning what a few hundred emails could have told you in weeks.

Cold outreach, run as GTM experiments

One question, one campaign, one verdict. The same loop works whether you're checking an ICP, an offer or demand for a feature.

Write the hypothesis

Segment, persona and offer, plus the reply rate that would make the channel pay off in your own economics.

Prove delivery

Set up domains and mailboxes, then confirm the emails reach inboxes before anything is read as a verdict.

Send to a fresh sample

Companies nobody has emailed, the same sequence for everyone, and a sample size decided in advance.

Read the verdict, then scale

Works, doesn't work, or too close to call. Only a “works” becomes a fully automated campaign.

Pick the question you need answered:

Validate your ICP Test an offer Check demand for a feature

What you'll learn

Five chapters, in the order an experiment runs: from proving delivery to scaling the campaign that earned it.

Chapter 1

Prove delivery before you test anything

Domains, DNS, reverse DNS, mailboxes and warmup, and how to confirm your emails actually arrive before you trust a result.

Chapter 2

Define a segment you can test

A narrow ICP with one shared problem, enough companies nobody has emailed, verified contacts and qualification against your hypothesis.

Chapter 3

Write one offer worth testing

A four-beat cold email that leads with their problem and ends with a real question, and why the copy stays fixed while you test.

Chapter 4

Run each campaign as an experiment

A threshold from your own economics, what an answer costs in emails, stopping rules decided up front, and the mistakes that fake a result.

Chapter 5

Scale what works

Turn a “works” verdict into an automated campaign: more mailboxes, healthy domains, replies, CRM and calls, and when to go back to testing.

Run it in Sellvance

Sellvance finds the companies, sends the cold emails from your own domains, checks delivery and tells you when the answer is in.

Start your first experiment

Your experiment checklist

Before you scale anything, make sure each of these is true. Every item is covered in the chapters above.

Delivery

Separate domains for outreach
SPF, DKIM, DMARC and reverse DNS verified
Mailboxes warmed up
Placement tested on the real message

Segment

One segment where everyone shares the problem
Enough matching companies for the sample
Previously contacted companies excluded
Emails verified and leads qualified

Experiment

Threshold set from your economics
Sample size and stopping rules written down
One offer, copy fixed for the whole test
Delivery read before replies

Scale

Only after a “works” verdict
A separate campaign with more mailboxes
Delivery and unsubscribes still watched
Replies routed to your CRM and calendar

Run your first GTM experiment

Validate your ICP, test an offer or check demand for a feature with cold outreach. Sellvance proves the emails arrived, tells you when the answer is in, and runs what works on autopilot.

Start your first experiment Read the Experiment Playbook