The Voice of Email Methodology
Four stages, in order, every time. Then two real engagements traced end to end. One cold outbound campaign for a sales team, one lifecycle program for a marketing team, each from the first thing I read to the number each produced.
And you should leave with something. Every stage below ends with the check you can run on your own program this week, whether or not we ever speak. If you use four of them and fix the problem yourself, that's a good outcome. It usually means you didn't have the problem I get hired for.
A prospect told me last year that a finished sample wouldn't impress him. His words: he wanted to know what the thing actually reads, what data goes into it, and where the judgment gets made. Only then would he consider buying.
He was right, and it's the fairest question anyone has asked me. A polished email proves someone can write. It proves nothing about whether the same result happens twice. So this page is the trace, not the sample.
Four stages. In that order, every time.
It starts where voice of customer research starts, in your buyers' actual words, and it does not stop until those words are in the inbox and on the scoreboard. Skip a stage and the next three carry that problem forward. Most programs start at stage three and wonder why the machine is loud and the pipeline is quiet.
The Read
Everything you have already sent, and everything underneath it. Prior campaigns, the list, the replies, the sending posture, your recorded sales calls. The point is to find out whether words are the problem before anyone is paid to write words. Two thirds of the time they are not.
Outbound: prior sends, the list, the sending posture, the replies.
Lifecycle: where each person came from, what they did next, what they bought, and why the ones who left, left.
The Decision
One working session, one page. Who it is for, what it is against, and the one claim only you can make, written in the words your buyers already use. Every email after this gets written from that page, which is the only thing that stops the next rewrite from reopening the argument.
Outbound: what a stranger needs to believe to answer at all.
Lifecycle: who the reader is trying to become. People sign up as a decision about themselves, and every email after that either confirms it or contradicts it.
The Send
Authentication, warmup, throttling and a live placement test first. Then the list split by relationship, with the group that has no history suppressed rather than burned. Then, and only then, the writing. The best sequence ever written cannot outrun a domain that inboxes have decided to distrust.
Outbound: split by relationship. Engaged, went quiet, never replied, and the third group gets suppressed.
Lifecycle: split by source and behavior, then build the flows that were never built: activation, nurture, winning people back. One idea and one ask per email.
The Scoreboard
Replies from real buyers and meetings booked on the top two rows. Opens demoted to what they always were, a deliverability signal. Plus the reply ladder, because the leak is almost never the email. It is the hour after somebody answers it.
Outbound: replies from real buyers, then meetings booked.
Lifecycle: activations, upgrades, and revenue traced to the sequence that caused them.
One live sequence and the decision page are in your hands by day 30. The documented playbook by day 90.
Every email program lands in one of three and they need different things.
Knowing which one you have is most of the value, and it is the part almost nobody establishes before spending money.
A plumbing problem
Authentication is broken, the domain is young or burned, the list was never verified. The words are irrelevant until this is fixed, and it is the cheapest of the three to solve.
A decision problem
Nobody has agreed what the company is. Three leaders describe three companies. Every rewrite reopens the argument, so no writer can rescue it. This is the expensive one, and it is the most common.
A copy problem
The plumbing works, the decision exists, and the writing still is not landing. Genuinely rare on its own, and the only one of the three that a copywriter alone can fix.
Voice of Email, running on a cold outbound campaign.
Karias Health, a benefits platform sold through brokers. Britt Prim, their Chief Experience Officer, has agreed to be named and will take a call about it. Every stage below ends with the check you can run on your own program today. The marketing side follows right after.
Britt was new in the seat. That matters more than it sounds: a leader in their first ninety days is the only person in a company with permission to say the current thing isn't working.
What she inherited, in her own framing, was a program that hadn't been doing much. So the first job was not writing. It was reading. I went through their prior sends, the broker list itself, and the sending posture underneath both.
What the read turned up
Open rate sitting at 14%. Low enough that no rewrite alone was going to move it.
A send ceiling. Somewhere between 30 and 75 messages, the sending had been getting flagged. That is an infrastructure fact, not a copy fact, and it caps everything downstream.
A broker list nobody had segmented. Engaged, gone quiet, and never-replied were all getting the same treatment.
Two of those three findings had nothing to do with words. If I had shown up and rewritten the emails, the new ones would have gone to the same place the old ones went.
Run this yourself, today
Pull three numbers before you let anyone touch the copy:
- Your last campaign's open rate, and the send volume it ran at.
- The send count where things started degrading. Most teams have a ceiling and have never gone looking for it.
- Whether your SPF, DKIM, and DMARC records actually exist and align on the domain you send from. Not the domain your website is on. The one the emails leave from.
If you can't produce all three inside an hour, that gap is itself the finding, and it's the cheapest one you'll ever fix.
This is the part that gets skipped, and it's the part that does the work. Not the subject line. Not the call to action. The question of what this company is, to this reader, in a sentence they'd repeat.
There's a piece of evidence I trust more than any brief here, and it came from how Britt and I met. I sent her a message that offered to compare notes on what was working in healthcare. It didn't pitch anything. It didn't mention what I sell. It treated her as a peer with information worth trading.
Britt, later, to her own team
“Ben sent me a cold outreach message on LinkedIn right when I started at Karias. He's the only person I have ever in my entire life responded to.”
That frame became the spine of the campaign. The brokers weren't going to be sold a platform in a cold email. They were going to be handed something a broker would find useful, in a broker's own vocabulary, by someone who clearly knew what their week looked like.
The decision took one working session. It went on one page. Every email after that got written from it, which is the only reason the argument about what to say never restarted.
Run this yourself, today
The one sentence test. Ask three people on your leadership team, separately and in writing so they can't hear each other:
“What is this company, in one sentence, to someone who has never heard of us?”
Then ask one more question: “When a happy customer describes us to a peer, what do they call us? In their words, not ours.”
I have run this on calls where three leaders produced three different companies. If that's your result, no rewrite fixes it, because there is nothing agreed on for the writing to be built from. That is a thirty minute exercise and it will tell you more than a quarter of A/B tests.
The list was 729 brokers. Before any of them got an email, two things happened that most programs skip entirely.
First, the sending got fixed
The ceiling from stage one gets addressed before volume goes anywhere near it. Authentication, warmup, throttling, and a real placement check rather than a green dashboard light. There is no version of this where the writing outruns a domain that inboxes have decided to distrust.
Second, the list got split by relationship
Engaged, went quiet, never replied. Three groups with three different histories, which means three different messages, and one of them shouldn't be contacted at all yet. Blasting all three identically is the most common reason a reply rate reads like static.
Then the email itself. It opened on what a broker was already dealing with that quarter, in the words brokers use for it, and it asked for something small. It did not lead with the platform. A cold email that opens with what you sell has already told the reader it is about you.
Under 100 words. One ask. No attachment, no deck, no calendar link buried in a signature.
Run this yourself, today
Before your next campaign goes out, do these two things in this order:
- Split the list three ways by what the relationship actually is, and suppress the group with no history rather than burning it. Your reply rate is partly a list problem wearing a copy costume.
- Read your first email out loud and mark every sentence that is about you rather than about the reader. If the first two sentences are both about you, that's the rewrite, and you just did it for free.
open rate, before and after, same broker audience
click rate on a cold list that had ignored everything prior
brokers reached, replies back from real humans
lunch and learn, plus multiple first meetings and pipeline activity
Eleven replies out of 729 is the number a lot of people would quietly leave out. I'd rather show it, because it's the honest shape of cold outbound to a skeptical professional audience, and because those eleven turned into meetings and a room full of brokers rather than into a spreadsheet of opens.
Britt Prim · Chief Experience Officer, Karias Health
“Ben is so good at what he does that he is the first and only cold outreach I have ever responded to over the course of my 20 year career. He absolutely overdelivered on what I was expecting. His campaign drove multiple first meetings, a lunch and learn, and pipeline activity. 10/10 would recommend.”
“The difference between you and an agency — you're giving full dedicated time and effort versus handoffs, pass-offs, overhead.”
Run this yourself, today
Look at the top line of your last email report. If it leads with opens, you are grading the wrong thing, and you have been for a while. Rebuild it so the first two rows are replies from real buyers and meetings booked, and let opens sit underneath as a deliverability signal, which is the only thing they were ever good for.
Do that one thing and half the arguments about copy in your company will end, because the scoreboard will finally be measuring the thing everyone claims to care about.
The same four stages, on a lifecycle program.
Cold outbound is the loud half of this job. It is not the bigger half. Sycle is audiology practice management software with thousands of clinics already in the database, which means arrival was never the problem. Earning the next action was. Same four stages, completely different program.
Nothing here was broken in the way a cold program breaks. Authentication was fine. The list was thousands of real customers who had chosen to be there. The emails went out on schedule and people got them.
They were also functional and generic. They opened by talking about the product. And the real job, as their marketing leadership described it, was not a campaign problem at all: the market had started to think of them as behind the times, and every email was either fixing that or feeding it.
What the read turned up
A perception problem wearing a campaign costume. No amount of tuning the send volume addresses that.
Emails that opened on features rather than on the thing a practice manager was already worried about that week.
A tier migration coming that everyone internally was framing as a price increase, which is the framing most likely to make customers leave.
Run this yourself, today
Pull your last five lifecycle emails and count how many open with a sentence about you. Your product, your release, your webinar, your news. Then count how many open with something the reader was already thinking about. If the ratio is worse than one to one, you have found this week's work and it cost you ten minutes.
This is the part of lifecycle email almost everyone gets backwards, and it is worth being blunt about.
The instinct is to send value. Here is a tip, here is a guide, here is a webinar. Value emails are the most stereotypical emails in B2B and they mostly do nothing, because they treat the reader as someone collecting information.
The moment somebody joins your list, they have made a decision about who they want to become. A better operator. A practice that runs cleanly. The person whose team is not drowning. Every email after that either moves them toward that person or quietly tells them it was not for them.
So the decision here was not a subject line strategy. It was that the tier migration is not a price change, it is a decision about what kind of practice you are running. Reframed that way, an upgrade stops being a cost to resent and becomes a thing a certain kind of operator does.
Run this yourself, today
Before you write your next lifecycle email, finish this sentence in writing:
“This is the kind of [their job title] who ______.”
Then write the email to that person instead of to a segment. If you cannot finish the sentence, that is the actual finding, and it is the same finding as the one sentence test in trace one, arriving from the other direction.
What shipped, all written from the same page:
Partner campaigns written in the partner's voice rather than in the company's.
Product launch emails that opened on the workflow problem the release solved, not on the release.
Webinar sequences with subject lines specific enough that only an audiology practice would care, which is the entire point.
The tier pricing migration, rebuilt around the growth decision rather than the invoice.
Landing pages matched to each campaign, because a click into a page that says something different is a click you paid for and then wasted.
One idea per email. One ask per email. The rule is boring and almost nobody follows it, which is why it keeps working.
Run this yourself, today
Open your best performing lifecycle email and count the asks. Book a demo, read the blog, follow us, reply, register. If there is more than one, you have your explanation, and the fix is deleting rather than writing. Then click your own CTA and read the page it lands on as if you had never seen it before.
higher email click through rate across the program
revenue increase, from the campaigns alone
programs built: partner, launch, webinar, tier migration, pages
clinics on the platform the program had to speak to
An eight percent revenue increase traced to campaigns is a different kind of claim than a reply rate. It is also the only kind that survives a conversation with a CFO.
Sycle · Audiology Practice Management
“Ben has changed the game for our demand gen emails. He quickly learned our industry, our customers and prospects, and our needs. He delivers emails with a strong CTA which has led to higher clicks, opens, and deals.”
Run this yourself, today
Pick one lifecycle flow and trace revenue to it. Not attribution across the whole program, which nobody agrees on anyway. One flow, one cohort, one number. If you cannot do it for a single flow, that is why lifecycle email keeps losing budget arguments inside your company, and it is fixable in an afternoon.
Two programs with almost nothing in common. One was strangers who had never heard of the company. One was thousands of existing customers who already paid them money. Different lists, different channels, different definitions of success. Same four stages, in the same order, and in both cases the work that mattered happened before anybody wrote a sentence.
Four things that were true here and have been true every time since.
One engagement is an anecdote. These are the parts that have held across every program I've worked on since.
1. Two thirds of a copy problem is usually not copy.
Deliverability, list hygiene, and segmentation sit underneath every sentence you write. Fix the sentence on top of a broken layer and you have written a better message to a spam folder.
2. If your leadership team can't agree on one sentence, no writer can rescue you.
The decision has to exist before anybody writes from it. Without one, every rewrite reopens the same argument and every new hire relitigates it from scratch.
3. The first cold email is not where you sell.
It is where you prove you know what their week looks like. Selling is what the next four messages and the conversation are for. Almost every cold email that fails, fails by trying to close in message one.
4. The leak is usually the hour after the reply.
Programs that book meetings and programs that don't often have similar reply rates. What separates them is whether anyone knew what to send back to a yes, a maybe later, a referral, and an honest no.
The honest limits, since I've asked you to trust the rest.
One engagement is one engagement. Karias is the trace I can show completely, with a named executive who'll confirm it. It is not a promise that your numbers land in the same place. Different list, different market, different starting infrastructure.
This does not fix a product problem. If buyers understand exactly what you do and are choosing something else on the merits, that's a real answer and better email will only help you hear it faster.
If your message already works and you just need more of it, hire a systems shop that builds volume infrastructure. That's a genuinely different job and they're better at it than I am. I'll tell you that on the call rather than after the invoice.
And if the four checks above fix it, keep the money. That happens, and it's a fine outcome. It usually means the problem was mechanical, and mechanical problems don't need me.
How much of this I actually run.
companies whose entire sending fleet I pulled apart, domain by domain
specimen cold emails written from a real buyer's own public words
B2B companies since 2017, across healthcare, fintech, dev tools and vertical SaaS
inside HubSpot, Marketo, Customer.io, Instantly and the rest, hands on the keys
None of those are client outcomes. They're the amount of reps behind the judgment on this page, which is a different thing and worth knowing separately.
Run the four checks. Then tell me what you found.
Genuinely. If you do stages one through four on your own program this week, you'll know within an hour whether you have a copy problem, a plumbing problem, or a decision problem. Most people find it's the third one.
Send me what you got and I'll tell you which of the three it is, and whether you need me for it. That costs you an email.
Email me what you found →Or if you'd rather just talk, grab 15 minutes →
Britt Prim at Karias takes reference calls about this engagement. Ask and I'll connect you.