Reply rate benchmarks by industry — and what they hide

You're about to run a cold campaign. You want to know what a good reply rate looks like. You Google "cold email reply rate benchmarks" and find a dozen reports. SaaS averages 3%. Fintech 2%. Ecommerce 5%. The reports are tidy, sortable, and mostly meaningless without context.

Not because the numbers are wrong — most vendors compute them honestly off their own platform data. But each benchmark smooths over four distortions that determine whether you should expect to match it. Without controlling for these, you're comparing your campaign to someone else's entirely different campaign.

Distortion one: deliverability is not controlled for

The headline reply rate is always replies over messages sent. Sent, not inboxed. Two campaigns with identical copy and identical ICP can produce 1% and 5% reply rates just from a placement gap.

A vendor whose customer base averages 70% inbox placement will naturally report higher reply rates than a vendor whose customer base averages 40%. The difference has nothing to do with industry — it reflects whose senders are better warmed up. If you compare yourself to the 70% vendor's benchmark while running at 40% placement, you'll conclude your copy is bad. Your copy might be fine.

Distortion two: list quality is invisible

A 3% reply rate on a hand-curated 200-person list is a completely different signal from a 3% reply rate on a 20,000-person scraped list. The first is a boutique agency doing careful work. The second is a volume play with one bad reply for every cease-and-desist.

Benchmarks almost always aggregate both. They cannot separate "my prospect fit is good" from "my prospect fit is noise with volume." Your job when reading a benchmark is to ask: whose lists are in this sample? If the vendor caters to volume senders, the benchmark is volume-sender reality. If it caters to account-based outbound, the benchmark is narrower and higher quality.

Distortion three: reply definition varies

Some vendors count any reply. Some count unique senders. Some count positive vs. negative. Some strip auto-responders, OOOs, and bounce-backs. Some don't.

The difference between "any reply" and "positive reply" is typically 2–3x. A campaign at 4% any-reply might be 1.5% positive-reply. When comparing benchmarks from different vendors, assume you're comparing apples to oranges unless the methodology is spelled out.

Unsubscribe and "remove me"

Messages that say "take me off this list" count as replies in most vendor dashboards. If your sequence is aggressive, your reply rate will be inflated by negative replies. A 5% reply rate where 4% are "remove me" is actively damaging your domain reputation, even though the dashboard looks healthy.

Distortion four: survivorship bias

Vendors publish benchmarks from active customers. Customers who churned — often because their campaigns didn't work — are gone from the dataset. The surviving customers are, by selection, the ones whose campaigns are working well enough to justify the subscription.

This biases benchmarks upward. If you ran cold outreach for three months and got 0.5% reply, you're likely to cancel. Your data won't appear in the benchmark. The published "SaaS average 3%" is the average of survivors.

What the current benchmarks roughly look like

With all the caveats above, here are reasonable mid-range any-reply expectations for well-run cold outreach in late 2026. These assume authenticated sending, warmed domains, and at least 60% inbox placement.

• SaaS (horizontal): 2.5–4% any-reply. 0.8–1.5% positive-reply.
• Fintech: 1.5–3% any-reply. Prospects are more gatekept; legal/compliance scans a lot of inbound.
• Ecommerce brand services: 3–6% any-reply. Often higher because decision-makers are more reachable.
• Enterprise IT buyer: 1–2.5% any-reply. Strict filters, thick gatekeeping.
• Healthcare: 1–3% any-reply. Compliance-heavy, filter-heavy, often regulated correspondence rules.
• Agency/creative services: 2–4% any-reply.
• Dev tools selling to engineers: 2–4% any-reply when personalised well; crashes below 1% on generic outbound.

What the benchmarks can't tell you

1. Whether a given campaign will beat the benchmark. The benchmark is a population average. Your campaign is a single data point. Whether you'll match depends on deliverability, ICP fit, copy, and offer strength.
2. Whether the benchmark applies to your stage. A well-capitalised agency is running a different motion from a solo founder. Their benchmarks reflect their resources.
3. What "good" means for you. A 1% reply rate is terrible if your ACV is low and you need volume. It's great if your ACV is high and you're running targeted outbound.

A more useful comparison

Instead of comparing to industry benchmarks, compare to your own previous campaigns. A week-over-week trend on your own data controls for all four distortions automatically. You know your ICP, your copy, your list, your provider — any movement is signal.

Specifically track:

• Inbox placement % (at Gmail, Outlook, and any vertical-specific providers).
• Reply rate on inboxed messages (not sent).
• Positive-reply rate.
• Meetings booked per 100 sent.

When one of these moves, you have diagnostic signal. When it's stable, you have a baseline. Either way, it's more actionable than a vendor benchmark.

FAQ