You spent real time building your writing profile. You fed it your best work, refined the output, and eventually got to a place where the drafts sound like you. Then you pointed it at an email and got back something that reads like a blog post that lost a fight with a word count limit.
The problem isn't your voice. Your voice is fine. The problem is that a profile built to produce sustained argument doesn't know it's now trying to write something a person will read on their phone in eleven seconds while waiting for coffee.
Different Format, Different Contract
Blog content earns attention. The reader has already opted in — they clicked, they're settled, they're willing to follow you through a setup and into a payoff. You can build an argument arc. You can use a section header to signal what's coming. You can take three paragraphs to get somewhere interesting.
Email doesn't work like that.
Litmus research puts average email reading time at 11–14 seconds. More than half of all emails are opened on mobile. The reader hasn't opted in to a reading experience — they're triaging. The question isn't "is this good?" It's "should I keep going or delete it?"
That changes the architecture entirely. A blog post front-loads the hook and earns the payoff. An email front-loads the core message and the ask. Everything that comes after is either supporting or losing them.
This isn't a voice problem. It's a configuration problem. Your voice — your specific rhythm, your word choices, the personality that makes your writing recognizable — travels across formats just fine. What doesn't travel is the structural logic of how a blog profile is built to operate.
What a Blog Profile Carries That Hurts Email
A well-tuned blog profile does exactly what you trained it to do. It builds. It uses transitional language to move between ideas. It introduces context before the point. It writes sentences that develop — not because it's being verbose, but because that's what good long-form writing looks like.
Every one of those instincts fights the inbox.
The Boomerang study analyzed over 40 million emails and found peak response rates at 75–100 words — right around 51%. Emails between 75 and 125 words consistently outperform anything longer when your goal is a reply. Constant Contact's data shows a different curve for click-through: CTR peaks around 150–200 words, roughly 20 lines, and then drops. Below 50 words and above 500 words, both response rate and CTR fall off.
A blog writing profile doesn't know any of this. It's calibrated to produce depth. When you point it at an email prompt, it will over-explain the context, build toward the CTA instead of opening with it, and use transitional structure — "Here's why that matters," "To understand this, consider," — that reads as delay in an inbox but reads as craft in a long-form piece.
On mobile, that's not just less effective. Long scrolling blocks actively hurt engagement. The reader doesn't hit the CTA because they never reach it.
The Blowout Run That Made This Obvious
When we were tuning the email profile for Riterly's beta, we ran the same marketing site writing profile against an email prompt — same profile, just different input. The output token count came back at over 700 words. A clean email run from a properly configured profile came back at 254.
That's not a rounding error. That's a profile doing what it was trained to do — and producing something five times longer than it needed to be, at roughly double the cost.
The culprit was input method. Detailed topic notes, written in the same expansive way you'd brief a long-form piece, drove long-form output. Lean, constrained input — a single sentence brief, format instruction included — fixed it. The profile wasn't wrong. The instructions were.
That's the non-obvious part: input method is part of the profile. How you brief the model shapes what it produces as much as any style rule. An email profile needs to specify not just what the output should look like, but how it expects to receive a prompt.
What an Email Profile Actually Needs
An email profile isn't a shorter blog profile. It's a different set of rules, built around what email is actually trying to accomplish.
Here's what that looks like in practice.
Word count ceiling, explicit. Not "keep it concise" — that's not a rule, it's a suggestion. Set a hard ceiling. Under 125 words for emails where you want a reply. 150–200 words if the goal is a click-through. If you're walking someone through a value proposition and handling objections, 250+ may be warranted — but that's the exception, not the default, and even then, over 500 words starts costing you.
Single CTA, named upfront. A blog profile might naturally include multiple threads the reader can follow. An email with two asks usually gets zero responses. The profile should enforce one CTA, and it should appear early — not as a conclusion to an argument you built, but as the point.
No headers. Section headers signal long-form reading. In email, they break the conversational register and make the message feel like a document. The profile should explicitly prohibit them.
Subject line as part of the template. Subject lines truncate on mobile past 50 characters — closer to 30–40 characters is safer if most of your list is opening on phone. If the profile isn't generating the subject line with that constraint in mind, you're leaving one of the highest-leverage parts of the email outside the profile's control entirely. Keep subject line generation inside the template with an explicit character limit.
Vary by email type. A welcome series can run longer because the reader just opted in — engagement intent is high. A promotional email should stay under 125 words. A follow-up can be two or three sentences and nothing else. The profile should encode these as variants, not leave them up to whoever's writing the brief.
These aren't stylistic preferences. They're structural rules the profile has to enforce — because the format won't forgive you for ignoring them the way a blog reader might.
Why Your Sent Folder Won't Build This for You
When you set up a blog or editorial writing profile in Riterly, the Voice Workshop can sample from your existing writing. Your published posts are representative of what you want to produce more of. The feedback loop works because the source material reflects intentional craft.
Your email sent folder doesn't work the same way.
The emails you've already sent were shaped by whatever habits and constraints you were operating under at the time — not by deliberate thinking about what makes email effective. They might be too long. They might bury the ask. They might reflect exactly the failure modes you're trying to avoid. Sampling from them trains the profile to reproduce those patterns.
Email profiles need to be built from scratch, with explicit rules, not reverse-engineered from past behavior. The voice can carry over — that's the point. But the structural constraints have to be constructed deliberately, with the format's actual mechanics in mind.
Your blog profile already knows how to sound like you. Your email profile needs to know what email is for.
COMMENTS
Login or Register to post comments.