Cold email can feel broken when you’re judging it too early. Testing with just 100 contacts makes every open, reply, or hostile message look bigger than it is—and that noise hides the real signal. In our recent Semantic Mastery session, we showed why bigger samples change the game, how to structure directory outreach tests, and the simple rules we follow to decide when a campaign is actually working.
Table of Contents
Why 100 Contacts Is Not Enough
Small samples lie. With only 100 emails, a few replies or a handful of opens can swing your numbers wildly. We can see this in real numbers. One campaign ran with 100-ish contacts and showed a 47% open rate. That sounds okay, and it is okay, but you cannot trust it as a final result.
We often compare multiple campaigns that use the same sequence and the same sending accounts. One campaign showed a 56% open rate and another showed an 83% open rate, while both used the same sequence and had excellent email health scores. Why the difference? Randomness. A small sample size can make it hard to know if you have a real pattern or just noise.
When a campaign is small, the effect of one or two outliers becomes huge. Open rates will bounce. Reply rates will jump or drop. A hostile reply feels worse than it is because it colors your view of the whole test. But with larger numbers, extremes smooth out and you see the real trend.
What We Mean by “Send More”
We recommend sending at least 1,000 contacts before making a call on whether a sequence or list is working. That does not mean blasting 1,000 emails all at once from one account. It means building a larger test sample across accounts and time so you can trust the metrics.
- 100 contacts = tiny test. Use only for rough sanity checks.
- 1,000 contacts = enough data to see meaningful patterns.
- 10,000+ contacts = clear signals for scaling and segmentation.
We emphasize patience. A bigger sample takes time, but it saves time later because you avoid throwing out good sequences or doubling down on bad ones based on random swings.
Got SEO Questions? Get answers every week at 4pm ET at Hump Day Hangouts. Ask questions ahead of time, or live – just go to: https://semanticmastery.com/hdho (bookmark this!) 10+ years of insights given every week!
Get your checklist to help get better results with GBPs, faster.
Key Metrics to Watch
When you scale your samples, look at a few core metrics to judge performance. Keep them simple and consistent across tests.
- Open Rate — How many people open your email. In cold outreach, 40–60% is common depending on the list and subject line. We saw one campaign with a 56% open rate and another with 83% on the same sequence. That shows how variable this can be.
- Reply Rate — How many people reply. This is the real signal for interest. Even with decent open rates, reply rates can be low if the message does not match the reader’s intent.
- Conversion Rate — How many replies turn into the next step (call, sign-up, acceptance). This matters for ROI.
- Negative Replies — Hostile replies are normal. With cold outreach, you will see a few. Ignore rude replies and learn from constructive feedback.
- Account Health — Sender health, spam complaints, and reputation. We had accounts with health scores above 99% in one example, and they still experienced different open rates across lists. Good health helps, but it is not a guarantee.
How to Build a Reliable Test
Here is a simple step-by-step method we use for directory outreach.
- Warm up multiple sending accounts for at least a month. Send small volumes of normal emails first.
- Keep sending volume per account below 50 emails per day while warming and testing. This helps avoid flags and keeps reputation high.
- Create a list of at least 1,000 target contacts. Use good segmentation: by state, industry, company size, etc.
- Use the same sequence across the sample so you measure list quality first, then tweak sequences later.
- Run the campaign slowly across accounts and time so deliverability stays strong.
- Track opens, replies, conversions, and negative replies. Watch for patterns by segment.
This process gives us data we can trust. With enough volume, we can compare how different regions or industries respond. We can also test subject lines and the body copy in a controlled way.
Handling Hostile Replies
Cold email draws attention—sometimes unwanted attention. In the example we discussed, there were two hostile replies. That is normal.
- Don’t take hostile replies personally. We do not know these people. They may be having a bad day or misread the email.
- Do not engage with insults. A short, polite opt-out or no reply is fine.
- Log negative replies and try to detect patterns. If the same complaint appears repeatedly, fix the messaging or targeting.
Hostile replies do not mean your campaign is broken. If the rest of the metrics are okay, keep going and get more data.
Why Lists and Segments Matter
Even with the same sequence and the same accounts, different lists behave differently. In our example, contacts in Iowa and Tennessee had higher open rates than contacts from Florida and Missouri. We cannot always explain the difference. Sometimes it is just luck. Other times it reflects differences in the people on the list, their role, or local habits.
Segmentation helps you find pockets where your message works best. When you reach 1,000 contacts, divide them by useful labels:
- Location (state, city)
- Industry
- Role or title
- Company size
- Source quality (where you found the contact)
Then compare open and reply rates across these segments. You might find your message resonates with some groups far more than others. That lets you tailor the next sequence for better results.
Test One Thing at a Time
When you run tests, change only one variable at a time. If you swap subject lines and email body at once, you won’t know what moved the needle. Keep things simple:
- Test subject line A vs B with the same body and same list split in half.
- Test email body variations only after you’ve chosen a winning subject line.
- Test different lists with the same sequence to measure list quality.
With large samples, you can run clean A/B tests and see real winners. Small samples hide true differences.
How Long Should a Test Run?
A test should run long enough to collect 1,000 data points or until you hit a clear pattern that persists across several days. We usually let a campaign run several weeks. Cold outreach often gets replies over time, not just immediately.
Do not judge a campaign after a few days or just a hundred sends. Instead, gather a bigger set and then look for trends. The email that seemed to flop on day three might perform well across a larger list.
Common Mistakes and How to Avoid Them
Here are mistakes we see often and how we fix them.
- Relying on small samples: Expand the list. Aim for 1,000+ contacts for useful analytics.
- Changing too many things at once: Test one variable at a time so you know what works.
- Ignoring reply quality: Open rates can be okay while replies are poor. Focus on replies and conversions.
- Letting account health slip: Warm accounts and keep volume steady. Health above 99% helps, but it is not the whole story.
- Panic from negative replies: Expect a few. Ignore rudeness and learn from patterns.
Simple Checklist Before You Scale
- Warm sending accounts for at least one month.
- Keep daily sends per account under 50 during early tests.
- Assemble a list of 1,000+ contacts for a proper test.
- Use consistent sequences across the sample for list testing.
- Track opens, replies, conversions, and negative replies.
- Segment the list and compare performance across segments.
- Run A/B tests that change only one element at a time.
Real Quotes We Use as Guideposts
“100 contacts not enough. Send out a thousand contacts and now you'll have something that's a little bit more significant.”
“47% open rate is fine. That's a damn good open rate on cold email.”
We like to keep these lines in mind. They remind us to be patient, test more broadly, and not overreact to early results.
Conclusion
Cold email is a numbers game and a testing game. Small tests give early hints, but they do not give answers. If your directory outreach is not performing yet, do not assume the sequence is bad or the list is worthless. Expand your sample size. Send at least 1,000 contacts across warmed accounts and watch the patterns emerge.
Keep account health high, keep sends steady, and split your lists to spot where your message lands best. Expect some rough replies and ignore the noise. With enough data, you will see what truly works and where to scale.
FAQ
How many emails should we send before judging a campaign?
We recommend at least 1,000 contacts. That gives you a data set that reduces random swings and helps reveal real trends.
What open rate is good for cold email?
Open rates vary by list and subject line. In cold outreach, 40–60% is common. We have seen as high as 83% in some runs. A 47% open rate is fine and not a cause for panic.
How should we handle hostile replies?
Ignore rude or hostile replies. They happen. Log feedback that is useful and look for patterns. If many people complain about the same thing, fix the message or the targeting.
How long do we need to warm accounts?
Warm accounts for at least a month with regular, low-volume sending. Don’t ramp too fast. Keep daily sends under 50 per account during testing to protect reputation.
What should we test first: subject line or body copy?
Test subject lines first with the same body and list. Once you have a winning subject line, test body copy variations. Change one thing at a time.
What if two identical campaigns have very different open rates?
This happens. Differences can be due to list quality, timing, or just randomness. Larger samples will smooth differences and show the real trend.
Can we scale before we hit 1,000 contacts?
We do not recommend scaling from tiny tests. Wait until you have enough data to prove the pattern. Otherwise, you risk scaling a false positive.