Cold email personalisation at scale: how AI changes the equation

For years, every outbound team faced the same trade-off: personalisation or scale. You could write 10 genuinely great, research-backed emails in a day, or 1,000 template emails with a first name swapped in. The first approach produced results. The second produced unsubscribes.

AI has collapsed that constraint. Not by making templates smarter — but by making research instant.

The research problem

The real bottleneck in personalised outbound was never writing. It was the 35 minutes before you wrote anything: opening LinkedIn, checking Crunchbase, reading the company blog, scanning their job listings, checking if they'd raised funding recently, figuring out what their product actually does.

That research is what made the email good. The writing was 5 minutes. The research was the rest.

What's different now

Modern AI doesn't just autocomplete your emails. It reads the company's website, runs targeted searches across the web, detects hiring signals and funding activity, identifies their tech stack, and returns a structured intelligence report — in under 10 seconds.

The message you write from that report is different from the message you'd write from a LinkedIn profile alone. It references something real. The prospect can tell.

The personalisation that actually works

After analysing thousands of leads, the signals that produce replies aren't generic ("I noticed you work at [Company]") — they're specific:

• →"I saw you just opened a London office and posted 3 BDR roles last month"
• →"Your recent Series B and the expansion into APAC puts you exactly in the market where we see the most traction"
• →"You mentioned scaling challenges in your last interview — that's exactly what our customers say before they start using us"

These details come from public data. They were always available. They just required 30 minutes to find — until now.

The result

Teams using AI-assisted research consistently report reply rates 2–3× higher than generic outbound. Not because the AI is magic, but because research-backed emails are better emails. The AI just made doing the research economically viable at scale.

Personalisation isn't a nice-to-have. It's the whole point. And now there's no excuse for skipping it.