Most founders who tell me outbound doesn’t work have tried exactly one version of it: pitch the whole product to everybody, send as many emails as the tool allows, and hope volume does the rest.
Then replies don’t come, and the conclusion is that the channel is dead.
It isn’t. But the thing they’re trying to fix — the copy — is almost never what’s broken.
The numbers have genuinely got worse
This part is real, and it’s worth knowing before you blame yourself.
Belkins analysed 16.5 million cold emails. Reply rates fell from 6.8% in 2023 to 5.8% in 2024, and every single month of 2024 performed worse than the same month the year before. That’s six straight years of decline.
So if your outbound stopped working, some of that is the market. Inboxes are noisier, filters are stricter, and everyone bought the same three tools.
What that doesn’t explain is why some teams are still booking meetings in the same market. The difference isn’t writing talent.
The mistake underneath everything
Founders try to pitch their entire platform to every person who’ll listen.
I had a founder tell me recently that they serve education businesses, service businesses, tech companies and cleaning companies. That’s not a market — that’s four markets, and one message can’t speak to any of them.
When you pitch to everybody, you pitch to nobody. And when industry reply rates are already under 4%, generic messaging is a death sentence.
What that looks like in practice
I started working with a bootstrapped tech company earlier this year. Good product, founder’s a great guy. He’d hired an SDR from a big legacy tech company and after two months they hadn’t booked a single meeting.
It wasn’t really the rep’s fault. He came from a business with inbound flowing and a marketing team feeding him leads — closer to an order taker than a pipeline builder. Different job entirely.
But the deeper problem was the messaging. Every email was about the platform. The features. The automations. The API. Nothing about what the person reading it was actually dealing with that week.
We cut about 90% of it. Removed the noise, the wordy scripts, the feature dumps. One specific problem, one specific persona.
Less than two weeks later: eight qualified opportunities.
Not because the copy was clever. Not because we sent more — we sent less.
Outbound is a chain, not an email
Five links. If one breaks, the whole motion crumbles, and you’ll usually blame the wrong link.
1. Segmentation
One segment, one problem, one message. Doing only this puts you ahead of most of the market.
The data backs it hard. Hunter analysed 11 million emails: lists under 50 people got a 5.8% reply rate, while lists over 1,000 dropped below 2%. Nearly triple the reply rate from sending to fewer people.
The reason isn’t mysterious. A list of 40 forces you to know something about the 40. You can’t write a generic email to 40 people you’ve researched. You can easily write one to 4,000 you haven’t.
2. Infrastructure
Most founders are quietly torching their domain reputation because they’re following advice from a video made in 2022.
After Gmail and Yahoo’s bulk sender policy update in 2024, the accepted safe range is 30–50 emails per inbox per day. If you’re still pushing 100–150, you’re burning your domain to send emails that land in spam.
Outbound is a reputation game now. It wasn’t five years ago.
3. Messaging
Personalisation isn’t “I loved your website” or “saw you’ve been at XYZ for a few years.” That’s a mail merge with manners.
It’s understanding a real problem the person is living through right now. Generic AI personalisation reads exactly like what it is, and buyers have got very good at spotting it.
4. Multi-channel
If you’re relying on one channel, you’re behind. The teams winning are running email, LinkedIn, calling, DMs — together, on the same accounts.
ProfitOutreach data shows outreach across three or more channels increases engagement by 200–280%, and prospects are eight times more likely to respond when touched across three or more.
Three channels isn’t three times the work, either. Same list, same research, same problem — different surfaces.
5. Qualification
More meetings isn’t the goal. More qualified meetings is.
If your system filters nothing, your calendar fills with people who were browsing. That feels like progress for about three weeks.
The founder’s trap
Founders lie to themselves about this, and I say that with love.
Ask how outbound’s going and you’ll hear: we’re doing enough outreach, we know our ICP, we’re multi-threading, our messaging is solid.
Look under the hood and the ICP is a vibe rather than a definition. The messaging is fluffy. There’s no channel mix. There are no buying triggers — nobody can say why this company, this week. The website still positions a platform instead of a problem. And the pipeline is full of fence sitters.
None of that is an effort problem. That’s what makes it hard to watch. These are people working genuinely hard.
Outbound doesn’t die from lack of effort. It dies from misalignment.
How to rebuild it in 30 days
If I had to start a motion from zero, this is the order I’d do it in.
Week 1 — Foundations. Pick one or two segments. Fix your sending infrastructure — domains, inboxes, warmup, limits. Write trigger-led messaging: why this company, why now.
Week 2 — Launch. Start small, 20–30 emails a day. You’re chasing signal, not volume. Read every reply properly, especially the negative ones — they tell you which assumption is wrong.
Week 3 — Optimise. Double down on the angles that got responses. Kill the ones that didn’t, even the ones you liked. Layer in calling and LinkedIn.
Week 4 — Pipeline. First meetings land — because you stopped talking about your platform and started talking about their problem.
Week 1 is the hardest and least satisfying part. There’s no output. It’s you and a spreadsheet arguing with yourself about who you actually sell to. Every team that skips it ends up back there in month four, having spent a quarter finding out the expensive way.
None of this is complicated. There are just more moving parts than anyone admits, and each one can quietly break the others.
It’s not rocket science. It’s doing the boring part first.