Which industries actually reply to cold email (the answer depends on who's asking)
Food & Beverage replies at 3.47%, about 8x the 0.45% average, in a 7.5M-email 2025 study. Banking and Insurance sit at the other end. Here's why.
By Gabriel Lara, Developer Relations, Norbelys
Founder-reviewed ·How we research and correct articles
Ask five cold email tools “which industry replies best” and you’ll get five different answers, usually the industry that vendor happens to sell into. That’s not necessarily dishonest — different agencies see different slices of the market, with different denominators, different list quality, and different definitions of “reply.” But one dataset is worth looking at closely because it’s internally consistent: Belkins’ 2025 analysis of 7.5 million of its own client emails, and the number at the top of it is not the industry most people would guess.
The actual number
Food & Beverage companies replied to cold email at 3.47% in 2025, against a 0.45% overall average across Belkins’ full dataset — roughly 8x the baseline. Education and Government both land clearly above average too. Banking and Insurance sit at the other end, described in the report as facing “structural headwinds” that hold reply rates down regardless of how good the email is.
| 2025 reply rate vs. the 0.45% average | |
|---|---|
| Food & Beverage | 3.47% — roughly 8x the overall average |
| Education | Above average |
| Government | Above average |
| Overall (all industries) | 0.45% (the 2025 benchmark) |
| Banking | Below average — "structural headwinds" |
| Insurance | Below average — "structural headwinds" |
Not every industry in that list gets a published decimal, and this post isn’t going to invent one to make the table look tidier. The direction is the real finding here: a food-service operator, a school district, or a government procurement office is meaningfully more likely to answer a cold email than a bank or an insurer is, and that gap held across 7.5 million sends.
Why food, education, and government outperform
None of these three sectors are obvious “high-tech, fast-moving” buyers — which is exactly the point. They’re comparatively under-targeted by the flood of SaaS and agency outbound that saturates tech-adjacent inboxes, so a well-timed pitch competes against less noise. They also tend to run on real, recurring procurement cycles — a school district budgeting for next year, a government office running a defined RFP process, a restaurant group sourcing new suppliers on a schedule — which means an unsolicited email sometimes lands exactly when the recipient already has a live reason to be looking. That’s a structurally different situation from cold-emailing a VP of Engineering who gets pitched a dozen dev tools a week whether they’re buying or not.
Why banking and insurance lag
The flip side is regulatory and cultural, not a copywriting problem. Banking and insurance run on formal vendor-approval processes, compliance review, and a general institutional wariness about engaging with anything that looks like an unsolicited sales pitch, because doing so can create real procedural headaches for the person who replies. No first line fixes that. It’s a structural ceiling on reply rate that sits below what the same recipient, in a less-regulated industry, would produce for an identically written email.
Where SaaS and software actually sit
Software is conspicuously absent from the top of every version of this list, and that tracks with the broader compression of reply rates over the past several years — SaaS is the single most heavily prospected category in B2B outbound, which means the same saturation effect working in Food & Beverage’s favor works against it. The 2026 benchmark breakdown covers what still moves the number inside a saturated category: personalization depth does more work there than in a less-targeted vertical, because it has to overcome more competing noise to be noticed at all.
The other variable that compounds with industry: company size
Industry isn’t the only segment Belkins’ 2025 dataset breaks reply rate down by. Company size moves the number too, and in the same direction you’d expect from the saturation story above: smaller companies replied at 0.49% (11-50 employees), while the largest enterprises, 10,000+ employees, came in at 0.22% — less than half. A small food-service company and a 10,000-person bank represent two ends of the same underlying mechanism from two different angles: the food-service company is under-targeted because of its industry, the small company is under-targeted because of its size, and a large enterprise in an under-targeted industry still has enough internal gatekeeping and inbound noise of its own to pull the number back down. Industry and company size aren’t independent variables — they compound, which is worth knowing before you conclude a below-average number in one dimension is the whole story.
What to actually do with this
Industry isn’t destiny and it isn’t an excuse. A below-average-reply-rate sector doesn’t mean stop sending — Banking and Insurance are enormous addressable markets, and even a below-average reply rate on a few hundred thousand well-targeted sends is still real pipeline. What the data is actually good for is calibration: if your program targets banks and insurers and your reply rate sits meaningfully below your Food & Beverage peers’ numbers, that’s expected, not evidence your copy or your list is broken. If your program targets Education or Government and you’re pulling numbers close to the SaaS-saturated average, that’s the more useful signal — something in the list, the offer, or the timing is probably leaving replies on the table that the vertical would otherwise support.
The only way to know which situation you’re in is to segment your own reporting by industry instead of reading one blended number. Norbelys’s audience segments build on the actual industry field on each contact record, so you can define “Banking” or “Food & Beverage” as a live, dynamic segment and watch its reply rate independently instead of guessing whether an industry mix is dragging your average up or down. Pair that with human-verified analytics — filtered for the bot opens and scanner clicks that inflate raw pixel counts — and you get a reply-rate baseline you can actually trust industry by industry, not just in aggregate.
If you’re running outbound into more than one vertical right now, the fastest way to find your own version of this data is to split your reporting by industry this week. See what’s included on every Norbelys plan and start segmenting your audience today.