No Replies to Your Cold Emails? The 9 Causes, in the Order to Check Them (2026)
The sequence went out, the dashboard shows opens, and nobody answers. Most teams start rewriting the copy, which is usually the wrong end. The diagnostic we run in that order: deliverability, list, timing, ask, sequence, sender, and only then the words.
Key takeaways
- Silence is a symptom with an order of likely causes. In most audits we run, the problem sits in delivery, list or timing, not in the copy the team spent a week rewriting.
- Check in this order: are the emails landing, are they reaching the right people, is there a reason to write now, is the ask small enough, is there a sequence, who is the sender, and only then the words.
- Under about 200 sends per variant, you cannot tell a bad message from bad luck. Fix the measurable causes first, then judge the copy on a sample that means something.
- The German-speaking market punishes volume and rewards specificity: a visible trigger, a one-line ask and a real sender beat any subject-line trick.
Every week someone shows us the same dashboard: a few hundred emails sent, a healthy-looking open rate, and a reply column that reads zero. The reflex is to rewrite the message. In most of the audits we run, the message is not the first problem, and often not the second either. Here are the nine causes we check, in the order that finds the real one fastest.
The direct answer
No replies almost always come from one of nine causes, and the order matters because the early ones make the later ones invisible. If your emails land in spam, no copy on earth will get a reply. If they reach the wrong people, timing does not matter. So check delivery first, then the list, then the trigger, then the ask, and treat the words as the last suspect rather than the first. The good news is that the early causes are the easiest to measure and to fix.
Cause 1: the emails are not landing
Open rates are not proof of delivery any more; mail clients pre-load images and inflate them. Test placement directly: send the exact sequence to a handful of seed inboxes at Google, Microsoft and a German provider, and see where it lands. The usual culprits are a young domain without warmup, missing or misaligned SPF, DKIM and DMARC records, sending volume per inbox that jumped instead of ramping, a shared sending domain that someone else burned, and links or tracking pixels that spam filters dislike. If placement is broken, stop everything else and fix it; our deliverability guide covers the full setup.
Cause 2: the list is wrong
The second most common cause hides behind a clean bounce rate. The emails arrive, but at people who cannot answer: the wrong role, a title that changed a year ago, a company that no longer fits, a generic info address. Pull 30 rows at random and check them by hand on the company website and a professional network. If more than a fifth are wrong, the list is the problem, and sending more of it makes things worse. In DACH the honest sources for company data are the public registers plus a professional network as the truth layer for roles; the company data guide maps them.
Cause 3: there is no reason to write now
A correct email to the right person at a random moment is still a random email. What lifts reply rates more than anything we have measured is a visible trigger: a job posting in the relevant department, a leadership change, a new site, a regulation with a deadline, a technology switch. If your first line could have been sent to any company on any day, the recipient knows it, and in the German-speaking market that reads as mass mailing. Rebuild the list around events, not attributes. The signals guide lists the ones that work here.
Cause 4: the ask is too big
"Do you have 30 minutes for a demo next week" asks a stranger to commit time to your agenda. A one-line question that can be answered from the phone in the elevator asks for nothing. "Is the migration to the new ERP still planned for this year, or has it moved?" gets replies. Demo requests get archived. Shrink the ask until answering costs less than ignoring.
Cause 5: the message is about you
Count the sentences about your company versus the sentences about theirs. Most silent sequences run three to one in favour of the sender. The recipient does not know you and has no reason to care what you are proud of. Lead with the observation about them, make the connection to a problem they plausibly have, ask the small question. Company description belongs in the signature.
Cause 6: there is no sequence
One email is not outreach. In the data we see, most replies arrive on the second to fourth touch, not the first, and a single message with no follow-up simply measures how many people had a free minute at that moment. Build three to five touches over three to five weeks, each adding something new: a different angle, a relevant reference, a question that is easier to answer. Never "just checking in". The follow-up playbook shows the rhythm and the wording.
Cause 7: the sender is a system, not a person
Emails from info@, sales@ or a marketing automation with a banner in the header get the treatment they deserve. The sender needs to be a real person with a real name, a plausible role and a signature the recipient could verify. That person also needs to actually read the replies, because the second-best outcome after a meeting is a reply that says "not me, talk to my colleague", and that only helps if someone acts on it within a day.
Cause 8: the volume is too small to judge
This is the cause nobody wants to hear. Fifty emails with two replies and fifty with zero are statistically the same sequence. Below roughly 200 sends per variant you are reading noise. If the first seven causes are fixed, resist the urge to rewrite after every quiet week. Let a variant reach a sample size, then decide.
Cause 9: the words
Now, and only now, the copy. What matters in German B2B outreach: a subject line that describes the content rather than teasing it, the trigger in the first sentence, the connection in the second, the question in the third, no exclamation marks, no "I hope this email finds you well", no jargon the recipient would not use, and the right form of address for the audience. The length that works is four to six sentences on a phone screen. Everything else is a test, not a rule; the cold email templates give you variants to start from.
The diagnostic table
| Symptom | Most likely cause | Test in one afternoon |
|---|---|---|
| Opens look fine, zero replies | Delivery to spam or promotions | Seed inboxes at Google, Microsoft and a German provider |
| Bounces low, replies zero | List: wrong roles or stale titles | Hand-check 30 random rows |
| Replies exist but all say "not relevant" | No trigger, wrong segment | Rebuild the list around one signal |
| Positive replies, no meetings | Ask too big, or no one following up within a day | Shrink the ask, set a reply SLA |
| One reply per 100, then silence | Single touch, no sequence | Add touches two to four with new substance |
| Good numbers in week one, collapse in week three | Sending volume ramped too fast | Cut volume per inbox, re-warm |
- Most likely cause
- Delivery to spam or promotions
- Test in one afternoon
- Seed inboxes at Google, Microsoft and a German provider
- Most likely cause
- List: wrong roles or stale titles
- Test in one afternoon
- Hand-check 30 random rows
- Most likely cause
- No trigger, wrong segment
- Test in one afternoon
- Rebuild the list around one signal
- Most likely cause
- Ask too big, or no one following up within a day
- Test in one afternoon
- Shrink the ask, set a reply SLA
- Most likely cause
- Single touch, no sequence
- Test in one afternoon
- Add touches two to four with new substance
- Most likely cause
- Sending volume ramped too fast
- Test in one afternoon
- Cut volume per inbox, re-warm
What this looks like when it is fixed
A working sequence in DACH is quieter than the American playbooks suggest: lower volume, a real trigger, a person as sender, a question instead of a demo request, three to five touches, and someone reading the replies every morning. On lists built that way, reply rates in the mid single digits are normal and a third or more of positive replies become conversations. If you would rather have that built and run than debug it yourself, that is what we do as a cold email agency in your own accounts, and the pricing page shows what it costs by market size.
Frequently asked questions
Why does nobody reply to my cold emails?
In the order most likely: the emails are not landing in the inbox, the list reaches the wrong people, there is no visible reason to write now, the ask is too big, the message is about the sender, there is no follow-up sequence, the sender is a generic address, the sample is too small to judge, and only then the wording. Check delivery and list before touching the copy.
What is a normal reply rate for cold email in Germany?
On researched lists with a real trigger and a person as sender, mid single digits is normal and high single digits is good. Generic lists with a demo request as the ask typically sit below two percent. The more useful number is positive replies per hundred sends, and then how many of those become a first conversation, which on a good setup is a third or more.
How can I tell whether my emails land in spam?
Not from open rates, which mail clients inflate by pre-loading images. Send the exact sequence to seed inboxes at Google, Microsoft and a German provider and look where it lands. Then check SPF, DKIM and DMARC alignment, domain age and warmup, and whether sending volume per inbox ramped gradually. Placement problems make every other fix invisible.
How many follow-ups should a cold email sequence have?
Three to five touches over three to five weeks, each adding new substance: a different angle, a relevant reference, an easier question. Most replies arrive on the second to fourth touch. A single email with no follow-up measures who happened to have a free minute, not whether the message works.