You've built your cold email list. You've written what you think is solid copy. You hit send on 50 emails and... crickets. Or worse - you get replies, but they're the wrong kind of replies. People are interested in the wrong thing, or your meeting rate is half what it should be.
The problem isn't that cold email doesn't work. It's that you tested nothing before scaling. Most people either blast their full list on day one, or they obsess over tiny tweaks that don't matter. Neither approach gets you to a working campaign.
Here's how to actually test a cold email campaign so you know what works before you spend time and money on it.
Start With a Real Baseline Number
Before you test anything, you need to know what "working" looks like for your specific business and audience. This isn't theoretical - it's actual expected performance.
For cold email, the baseline metrics that matter are:
- Open rate: 25-35% is typical for cold email to warm lists. If you're getting 15%, something is wrong with deliverability or your subject line is genuinely bad.
- Reply rate: 5-12% is the range for a solid cold email campaign. Below 3% means your copy or angle needs work. Above 15% usually means you're either getting low-quality replies or your list quality is exceptional.
- Meeting rate: 30-50% of positive replies should convert to meetings. If you're at 20%, your follow-up or qualification is weak. If you're at 80%+, you might be booking meetings that won't close.
Write these down before you test. Your test needs to hit or beat these numbers, or it's not ready to scale.
Test One Variable at a Time (Not Everything)
The most common mistake is testing five things simultaneously - a new subject line, a new opening, a new angle, a new CTA, and a different send time. When something works, you have no idea what actually worked.
Instead, run small tests on single variables:
- Subject line test: Send 100 emails with subject line A, 100 with subject line B. Measure open rate only.
- Opening line test: Keep everything else identical except the first 1-2 lines. Send 100 each. Measure reply rate.
- CTA test: Same email, different ask. Send to 100 each. Measure reply rate and reply quality.
- Sending day/time test: Same email, different send time. Send to 50 on Tuesday 9am, 50 on Thursday 2pm. Measure open and reply rate.
Each test should be at least 100 emails minimum. Anything smaller and you're just seeing noise.
Test Subject Lines First (It's Your Biggest Lever)
Your subject line controls open rate, which controls everything downstream. A bad subject line kills your whole campaign before the body copy matters.
Run a 100-person test on each subject line. Measure open rate. A solid opener should hit 28%+ open rate on your second+ touches. On first touches to cold lists, expect 20-28%.
Here are two real subject line structures that consistently work:
Quick question about [specific thing they care about]
This works because it's genuinely low-pressure. It's specific, not generic. If you're emailing e-commerce brands, it's "Quick question about your Shopify checkout flow" - not just "Quick question."
Saw you [recent, verifiable action]
This works because it proves you actually looked at their company. "Saw you just closed a Series B" or "Saw you launched your new marketplace." It's specific, personalized, and signals you're not blasting templates.
Test one of these against your current subject line. Record the open rate. Whatever wins by 10%+ is your new control.
Test Your Angle Second (This Controls Reply Quality)
Once you have a subject line that gets opens, test whether people actually want to talk to you.
The angle is the "why I'm emailing you" - the problem you're solving or the thing you noticed about their business. A poor angle generates replies like "not interested" or "not relevant." A good angle generates curious, qualified replies.
Run a 100-person test with angle A versus 100 with angle B. Keep the opening, body, and CTA identical. Only change the angle statement.
Example: If you're a conversion rate optimization consultant:
Most agencies we talk to spend 60% of their monthly revenue on paid ads, but 70% of that traffic bounces before booking a demo. We've helped places like [example] cut that bounce rate by 40%, which cut their CAC in half.
Compare this to a vaguer angle like "We help agencies improve their websites." The first one is specific, problem-focused, and credible. The second one will get lower reply rates and lower-quality replies.
Track reply rate and reply quality (how many say "yes" versus "maybe" versus "sounds interesting but busy now"). Weaker angles generate more "maybe" replies. Strong angles generate direct yes or no.
Test Your CTA on the Second Follow-Up
Most people test their main email and call it a day. Your follow-up sequence matters more than your opening email - follow-ups get 40-60% of your replies on good campaigns.
Once you know your main email works, test CTAs on the second follow-up. Some angles work better with aggressive asks ("Let's grab 15 minutes"), while others work better with low-pressure asks ("Any thoughts?").
Send 100 follow-ups with a direct meeting ask, and 100 with a softer ask. Measure reply rate and meeting conversion rate on each.
Know Your Minimum Sample Size
100 emails per test variation is the practical minimum. At 100 emails, you need roughly 3-4% difference in metrics to be confident it's real and not random chance.
If you test with only 50 emails per variation, your results are basically noise. You'll think things work when they don't, or miss wins because the sample was too small.
If you're running an initial campaign and trying to find any working angle, start with 300-500 emails split across 3-5 variations. This gives you room to find what resonates without committing your whole list.
Track Everything In a Simple Spreadsheet
Use one row per test. Record:
- Test name ("Subject line test - variant A")
- Send date
- Number of emails sent
- Opens
- Open rate %
- Replies
- Reply rate %
- Positive replies
- Meetings booked
- Winner/loser
You want to track cold email campaigns systematically so you can actually see what's working across time. Most people skip this step and keep results in their head, which guarantees they'll make decisions based on gut feeling instead of data.
Scale Only When You Hit Your Baseline
Once a test hits your baseline metrics (or beats them), you scale it. "Scale" means sending to 500-1000 of your list, not your entire list.
If your test got 8% reply rate and you need 6% to be profitable, scale to 500 more people with that exact setup. Watch what happens at scale - sometimes things that work at 100 emails flatten out when you go to 500.
Only after 500 emails at solid performance should you consider this "proven" and use it for your main list push.
The Gap Between Knowing This and Running It
Testing a cold email campaign correctly takes discipline - you need to resist the urge to change things halfway through, manage multiple variations simultaneously, and wait long enough for data to be real. You also need reliable email infrastructure that doesn't throttle you, list quality that's actually good enough to test on, and copy that's at least baseline competent to begin with.
If you want to run all of this yourself, you can. If the testing, tracking, iteration, and list management feels like it's eating time you'd rather spend elsewhere, that's exactly the gap we fill at BEC Growth. We handle the infrastructure, the testing framework, the tracking, and the iteration so you see which campaigns actually work without running it yourself.
Related Guides
- How to Write Cold Emails That Actually Get Replies
- How to Track Cold Email Campaigns (So You Actually Know What's Working)
- How to Book 20 Meetings a Month with Cold Email (Without Losing Your Mind)
- How to Keep Your Cold Emails Out of the Spam Folder
- Cold Email Frequency: How Many Should You Actually Send?