What Is a Good Cold Email Reply Rate for Local Business Outreach?
· 4 min read
Search for a good cold email reply rate and you will find numbers from 2% to 10% and more. They are not wrong. They count different things, over different lists, and almost all of them are published by companies that sell sending tools. Before you compare yourself with any of them, find out what was counted.
Why the numbers disagree
Take three reasonable sources. One vendor says a good figure for broad B2B campaigns is between 3% and 6%. Another puts a well-targeted campaign on warmed-up mailboxes with a verified list at 2% to 4%. Others publish higher ranges again. The gaps come from five choices:
- What is divided. Replies divided by emails sent, by emails delivered, or by leads reached give different results.
- Which replies count. Out-of-office messages, "please remove me" and "who are you?" are replies, but not what you wanted.
- Who is on the list. Executives at large companies answer less than a plumber who reads his phone between jobs.
- How the list was made. Verified and hand-picked is different from bought and bulk.
- Who wrote it. A benchmark from a sending tool tends to flatter the tool.
The three numbers to track
For your own outreach, count these, per week:
| Number | How to get it | Why |
|---|---|---|
| Delivered | Sent, minus bounces | The real number of people who could answer |
| Reply rate | Unique people who replied ÷ delivered | Whether anyone reads you |
| Positive reply rate | People who showed interest ÷ delivered | Whether the offer fits |
Count each person once, not each message. Ignore open rates: opens are tracked with a hidden image that many mail apps block or pre-load, so the figure says little, and tracking it can hurt delivery.
A sample is not a rate
With thirty emails, one reply is 3%, and zero is 0%. Neither means much. Wait for a hundred or more delivered before you read anything into the number, and compare weeks only when you changed one thing.
Local businesses are a different audience
The advice in most benchmark posts comes from selling software to companies with a sales team and a company inbox. A local business is different:
- The owner often reads messages on a phone, in bursts.
- Many answer on WhatsApp or by phone, not by email, so an email reply rate alone undercounts: email, WhatsApp or a phone call.
- A shared inbox such as
info@may be read by someone else, or rarely. - A "reply" may be "call me on this number", which is a good one.
Count replies from every channel you use, against everyone you contacted.
What moves the number
These do, in roughly this order:
- The list. Right trade, a real gap, a working address. How to qualify a lead before you contact it.
- The first line. One real detail from their listing, not "I hope this finds you well". A cold email template for local business owners.
- The size of the ask. "Would it help if I sent an example?" is easier to answer than "Can we book a call?".
- Timing. What time of day to contact a local business.
- Follow-ups. Most replies come after the second or third contact: how to follow up without being annoying.
- Delivery. Mail in the spam folder cannot be answered: how many cold emails you can send per day.
Change one thing at a time. If the reply rate is low and the positive rate is lower, the list or the offer is wrong. If replies are fine and nobody is interested, the message is attracting the wrong people.
What to do with a reply
A reply is not the end. Sort it: interested, wants a price, has a question, not now, not interested, stop. Answer the first three quickly, and honour the last one at once: how to handle STOP and not interested replies.
Tracking without a spreadsheet
AyePin reads each reply from a known lead, sorts it (interested, wants to meet, asks the price, has a question, not now, not interested) and moves the lead along its pipeline, so replies from email and WhatsApp end up on the same lead. A request to stop closes the lead. What you do with the numbers is still yours to decide.
A short checklist
- Count delivered, reply rate and positive reply rate, once per person.
- Wait for about a hundred delivered before you judge.
- Count replies from every channel.
- Change one thing at a time.
- Treat any benchmark as a loose guide, and read what it counted.