Results posts are usually where outbound people show a screenshot of their best week and imply it's the average. I'd rather give you numbers with denominators, plus the context that makes them interpretable.
One note up front: I don't publish client names or identifying details without permission, so campaigns here are described by category. The mechanics are the useful part anyway.
The scale
30+ clients across 16+ industries, including four lead-generation agencies where I was the outbound engineer running campaigns on behalf of their clients. 17 campaigns are documented in detail on the case studies page, with the data source, the angle, and the reason each one worked.
A campaign with the full math
The clearest example I can share in full is a lending client targeting a competitor's borrowers. County recorder data identified everyone who took a loan from a direct competitor in the previous year. Each borrowing entity was resolved into a real decision-maker. Each email opened with a verifiable public fact about that person's own transaction, followed by a same-day quote offer.
Across four iterations: 468,000 emails sent, 384 positive replies, with positives running at 15-20% of all replies.
Why I'm giving you all three numbers: the raw positive count means nothing without the send volume, and the send volume means nothing without the positive-to-total-reply ratio. That last number is the one that indicates message quality. It says that when people replied, one in five to one in six was actually interested rather than annoyed. In a category where most cold email gets a rounding error of genuine interest, that ratio is the result. The 468,000 just tells you the machine ran.
Results that aren't reply rates
Some of the work I'm proudest of doesn't produce a marketing-friendly number.
- A team brought in eight new clients in two months while never touching the campaigns themselves. That's their founder's words, on my homepage with his name attached. For an agency, "we didn't have to handle the campaigns at all" is the deliverable.
- 75 qualified leads from a single competitor's LinkedIn follower list, with zero invented personalization. The intent signal was already in the list construction.
- A 23,389-lead upload verified at 23,387 delivered, with 2 blocked server-side, caught by exporting the campaign and comparing normalized emails rather than trusting the API's success response. An earlier attempt had silently dropped 16,000 leads because of one inverted flag. Nobody puts "the list was actually the list" on a results page, and it's the difference between a campaign that underperforms mysteriously and one you can reason about.
- A measured provider waterfall from a 5,000-contact head-to-head test across six vendors, which now determines enrichment order for every list I build. That test converts directly into cost saved per campaign, forever.
What the range actually looks like
Here's the part most people won't put in writing. In a daily ranking I run across live campaigns, the top performers on a given day typically land between roughly 0.2% and 0.7% positive reply rate against that day's sends, where positive means specifically interested, meeting booked, meeting request, or information request, not "someone typed back."
If that sounds low relative to what you've been promised elsewhere, that's the point. At a few hundred to a few thousand sends a day, that rate produces a steady flow of real conversations. Anyone quoting you 5% positive reply rates on cold traffic is either measuring "replies" loosely, working a tiny hand-built list, or describing one exceptional week.
What moves that number, in order of impact: list source quality first, segmentation second, the specific signal in the first line third, deliverability fourth, and everything else a distant fifth.
What I measure and report
- Positive reply rate against sends, by segment, role, and variant, not just aggregate.
- Total reply rate, and the positive share of it, which is the message-quality signal.
- Bounce rate against verified sends, watched against thresholds rather than eyeballed.
- Deliverability health: warmup state, mailbox reputation, blacklist status, daily send consistency.
- Coverage: what share of the addressable market has been reached and when it's due for another rotation.
I built automated monitoring for most of this because manual checking doesn't survive contact with eight simultaneous clients. The tooling is described here.
What I won't claim
I won't promise a meeting count before seeing your offer and your market, because the offer determines more of the outcome than my execution does. I won't quote reply rates from one client's category as if they'll transfer to yours; local business lists and enterprise SaaS lists behave nothing alike. And I won't show you a screenshot without the denominator.
What I will tell you on a first call is whether I think outbound is even the right channel for what you're selling. Sometimes the answer is no, and that's a cheaper conversation to have in week zero than in month three.