I sent 48 cold emails in three days trying to get my site listed somewhere. Zero positive replies. Six bounces. Three offers to pay.
I'm writing this up because the "how to get backlinks" advice I read before starting was almost entirely wrong about what would happen, and the numbers are more useful than the advice.
The shape of the attempt
The site is a free browser game portal — no downloads, no accounts, no paywall. The target was any page that already lists similar sites, plus directories and teacher/parent resource pages.
48 emails over three days, sent from a normal Gmail account, each one written by hand, each referencing something specific on the recipient's page.
Results:
| Outcome | Count |
|---|---|
| Positive reply | 0 |
| Paid offer (guest post / listing fee) | 3 |
| Auto-reply / "we'll get back to you" | 1 |
| Hard bounce | 6 |
| Silence | 38 |
The bounces are three different problems
This was the most useful thing I learned, because I'd assumed "bounce" was one failure mode.
1. Address doesn't exist. Two cases. Fine, delete and move on.
2. The recipient's server rejects the sender. Two cases, and this one is worse than it looks. The message is not "no such user" — it's a spam verdict on me. A Gmail address sending near-identical cold mail to strangers gets filtered at the server level, and that reputation is shared across everything else I send from that account. Continuing to increase volume makes this worse, not better.
3. Their mail server is unreachable. Two cases. One domain's MX didn't answer on either IPv4 or IPv6 for a day. Nothing to fix on my side.
So: 12.5% bounce rate, and only a third of it is my fault in a way I can fix.
What the three paid offers cost
For the record, since it's useful for calibration:
- $65 for a "sponsored mention" in a listicle
- $149 for a guest post
- $50/month for a directory listing
All three were cheap, all three were from sites I would not want to be associated with, and all three would have been a nofollow link on a page full of other paid links. The fact that the only replies I got were sales pitches tells you what the inbox looks like from the other side: most people emailing site owners are selling something.
What I changed after day two
Instead of asking to be added to someone's list, I wrote the article first and offered that. A finished 800-word piece is a different proposition from "please consider my site" — one requires the recipient to do work, the other requires them to do nothing.
Response rate on that approach so far: also zero. But it's early, and the asymmetry is obvious enough that I'm keeping it.
The thing that actually worked
Publishing this post, on a platform that lets me link out without asking anyone's permission.
That's not a clever insight, but it reframes the problem: I spent three days on a channel where every single action required another person to say yes, and got 0 yeses. The channel where I needed nobody's permission took twenty minutes.
Two links in two days, one of them dofollow, versus zero after 48 emails.
What I'd tell someone starting this
- Count the yes-gates before you start. If every step requires a stranger's approval, your throughput is their inbox behaviour, not your effort.
- Check whether your bounce reason is about you. "Blocked" and "no such user" look similar in a mail client and mean completely different things.
- Do the free channel first. It costs an afternoon and it's the only one where the result is deterministic.
- Write down the stop condition. Mine is: if the reply rate is still zero after 30 days, the channel is wrong, not the volume.
The 48 emails weren't wasted — I learned what the inbox looks like, and I got a post out of it. But I'd have started with the writing.
I build GameSprig, a free browser game site with no downloads and no accounts.
Top comments (0)