Cold email subject lines that get opened

Part of our complete guide to B2B cold email.

The job of a subject line

A cold email subject line has one job: get the email opened. Not summarise your pitch. Not create urgency. Not sell. Open the email. Everything else is the body’s problem — for that, see our B2B cold email template library.

That framing alone eliminates 80% of bad subject-line instincts — the ALL CAPS, the fake “Re:”, the “LAST CHANCE” — all of which hurt open rates and deliverability at the same time.

What consistently works in 2026

Across thousands of UK B2B campaigns we’ve run and observed, the patterns that pull the highest open rates share four traits:

• Short. 3–7 words. Anything longer truncates on mobile and looks like marketing.
• Specific. References something concrete about the prospect or their business.
• Low-case. Lowercase subject lines outperform title case. They look like real emails, not campaigns.
• Question or fragment. Complete, polished sentences feel corporate. Incomplete ones feel human.

Why a swipe file is the wrong tool for this job

Here’s the problem with subject line lists — including this one. A subject line that pulls a 62% open rate in SaaS enterprise outreach may pull 28% in the same campaign targeting SMB trades businesses. The performance is context-dependent: it varies by ICP, by industry, by sequence position, and by what your prospects are seeing from every other person emailing them that week.

The right subject line for your audience isn’t in a list — it’s in your data. The list below is a starting inventory of proven patterns. The job is to pick 2–3 candidates, test them against each other at statistically meaningful volume, and iterate. A subject line you tested and validated against your specific audience will always outperform a borrowed one, regardless of what its benchmark rate looks like.

We run this testing loop for every PrawnMail client, every week — varying subject lines across audience segments and sequence positions, then applying what actually moves the number. The patterns below are where that process starts, not where it ends.

Subject lines that work, by intent

Open-rate ranges below are from real UK B2B campaigns we’ve run or audited. Treat them as directional signals, not benchmarks — performance varies meaningfully by industry, ICP, and sequence context. The formulas explain why each pattern works; that understanding transfers more reliably than the specific wording.

Subject lines that will put you in spam

These patterns either trigger filters directly or cause the reply/complaint behaviour that triggers them over time.

The emoji point is nuanced — a single tasteful emoji can lift opens in the right context, but stacks of them scream “campaign” and lower deliverability. Default to none.

Testing subject lines properly

Most people pick a subject line from a list like this one, use it until the campaign ends, and never know whether a different choice would have performed better. That’s not testing — it’s guessing with extra steps. Real subject line optimisation requires running variants at statistically meaningful volume, and most small businesses running their own outreach never get there because the sample sizes required to get clean results take time to accumulate.

For reference, a 3% difference between two subject lines on a 100-person send is noise, not signal.

To test properly:

• Minimum 200 sends per variant. Below that, don’t trust the result.
• Test one variable at a time. If you change the opener and the subject line, you won’t know which moved the metric.
• Run for 5+ business days. Day-of-week heavily affects open rates.
• Segment your results. A subject line that works for SaaS VPs may not work for agency founders. Test within your actual ICP.

What to do next

Pick 3 subject lines from the categories above that fit your offer. Run them on a batch of 200–300 prospects per variant. Compare open rates after 5 business days. Keep the winner, kill the others, write two new challengers to test against the winner.

That’s the loop. Everyone who’s good at cold email runs it continuously. The specific subject lines here are a starting inventory, not a final answer.

What ongoing subject line management actually looks like

For context on how this works in practice: for each PrawnMail client we maintain a live subject line testing matrix. At any given time, two to three variants are running across the active sequence. After 5 business days and a minimum of 200 sends per variant, we rotate out the underperformer, promote the winner, and write two new challengers based on what the data showed.

We also segment by audience type within a campaign — the subject line that outperforms for a founder of a 3-person business is rarely the same one that outperforms for a Head of Operations at a 50-person firm, even within the same ICP. Collapsing that into a single subject line is one of the main reasons DIY cold email underperforms. The patterns above are where that process begins. They’re not a substitute for running it.