The Executive Voice Print

🎯 Expert Mode: The "Executive Voice Print"

Scenario: You ghostwrite for an executive, and every AI draft comes back generic. You need a reusable voice profile you can paste into any prompt, hand to any writer, and carry to any tool.

Instructions:

1️⃣ Gather 3 to 5 writing samples your executive actually wrote or heavily edited. Bylines, all-hands emails, LinkedIn posts, board memos. Interview transcripts count too, and they are often the most honest source you have.

2️⃣ Open a fresh chat and paste this:

You are a voice analyst. Below are writing samples from [NAME], [TITLE] at [COMPANY]. Build a Voice Print: a reusable profile another writer could use to draft in their voice.

Analyze for sentence length and rhythm, how they open a piece, vocabulary tics they reach for repeatedly, words and constructions they never use, pronoun habits ("I" versus "we" versus passive voice), how they handle uncertainty and bad news, how they assign credit, and any signature structures like lists, one-line paragraphs, or callbacks.

Output three things. First, 8 to 10 rules written as instructions a writer could follow. Second, 3 verbatim sentences from the samples that best exemplify the voice. Third, 3 counterfeit sentences you invent that sound close but wrong, with a line on what gives each one away.

Work only from what is in the samples. If a pattern appears only once, say so rather than promoting it to a rule.

[PASTE SAMPLES]

3️⃣ Read the rules out loud and cross out anything that describes good writing in general rather than this person specifically. "Uses clear language" is noise. "Opens with the objection before the argument" is signal.

4️⃣ Test it. Pick a topic they have never written about, paste the Voice Print and ask for 200 words. Send it to them without context and ask one question: Does this sound like you?

5️⃣ Save the surviving Voice Print somewhere you can reach in five seconds. Every future draft starts by pasting it in.

Suggested Tool: Claude (handles long samples without losing the thread, and it is strong at pattern extraction across multiple documents)

Why You'll Love It

It replaces "make it sound more executive" with rules a model can actually follow.
The approval loop gets shorter, because the first draft already lands closer to right.
It's portable. Hand it to a new writer, a new agency, or a new tool and the voice survives the handoff.

Voice is the last thing AI can't guess at, because it lives in the choices your executive makes without noticing. Write those choices down once, and you stop rebuilding them from scratch on every single draft.

Talk soon,

Chris

⚡ The Cowork Version

Same idea, except Claude reads the whole archive itself and leaves a finished profile in your folder.

Setup (one time):

Make a folder for your executive and drop in everything you have: bylines, all-hands emails, board memos, keynote transcripts, LinkedIn exports.
Include material you know a team ghostwrote. Claude needs the contrast to tell the real voice from the house voice.
Optional: connect Google Drive, Notion, or Slack so it can pull the rest without you having to download anything.

Run this Cowork command:

Build a Voice Print for [NAME], [TITLE] at [COMPANY], using the material in this folder.

1️⃣ Read every document and sort them into what [NAME] almost certainly wrote, what a team almost certainly drafted, and what you cannot tell. Show me your reasoning on the uncertain ones and ask before you use them.

2️⃣ Extract voice patterns from the confirmed samples only: sentence rhythm, openers, vocabulary tics, forbidden words, pronoun habits, how they handle bad news, how they assign credit.

3️⃣ Map where the voice shifts by context, internal versus external and crisis versus launch, and where it holds constant no matter the audience.

4️⃣ Write Voice Print - [NAME].md to this folder with 8 to 10 rules, 3 verbatim exemplar sentences, and 3 counterfeit sentences showing what close but wrong sounds like.

5️⃣ Draft a 200-word test piece on a topic they have never covered, using only the Voice Print, and save it as Voice Test - [NAME].md next to the profile so I can check your work.

6️⃣ Save a short Voice Print - Sources.md listing which documents you treated as authentic and which you set aside, with one line of reasoning each.

What you'll get:

A Voice Print - [NAME].md file that works in any tool, any chat, and any agency relationship you hand it to
A context map showing how the voice flexes between internal and external writing
A test draft ready for the "does this sound like you" gut check, written before you ask for anything real
A source log separating what your executive actually wrote from what was written for them, which is quietly useful on its own

Want it on autopilot? Ask Claude to save the Voice Print as a skill. Every future draft request pulls it in automatically, and you re-run the whole build once a quarter as new writing accumulates.

Talk soon,

Chris

Here are a few articles from the past week to help keep you up to speed:

1. How to Capture an Executive's Voice on the Page

Caliber

August 10, 2026

Aaron Enneking makes the case for building a documented voice profile before you ghostwrite a single word. The advice hits differently now that most first drafts start in a chat window. If you can't describe how your CEO actually sounds, the model will hand you back the average of every executive who has ever published a byline.

2. How Disclosure Language Influences Retail Investor Attention When AI Writes for AI to Read

CommPRO

August 10, 2026

Ruoyi Huang digs into what happens when corporate disclosures are drafted with AI and then consumed by AI on the other end. She calls it the "AI Loop." For IR and corporate comms teams, it's the first serious look at writing for two audiences at once, and the tradeoffs are not the ones you'd guess.

3. Crisis Communications In The Synthetic Media Age: Proof As Part Of The Message

Forbes Communications Council

August 6, 2026

The argument is blunt: a denial is no longer a response. When a convincing fake can be produced in minutes, your statement has to include verifiable proof, not sit next to it. Worth reading before your next tabletop exercise, because most crisis playbooks still quietly assume video counts as evidence.

4. KPMG Says Nearly Half Of Executives Pulled Back AI Agents Over Cost

Forbes

August 9, 2026

KPMG's Q2 Global AI Pulse found that nearly half of executives have scaled back agent deployments because costs have outpaced benefits. The comms read: the internal story about AI is about to shift from expansion to justification, and someone will be asked to write it. Probably you.

5. A new book looks at how AI is rewiring the newsroom, for better and worse

Nieman Journalism Lab

August 6, 2026

Seth C. Lewis reviews "Journalism in the Age of AI," which maps AI across five stages of news production: ideation, sourcing, verification, storytelling, and distribution. Read it as a map of how the reporter you're pitching is actually working right now.

Ready to Future-Proof Your PR Strategy with AI? Here’s How I Can Help:

AI Training & Implementation Workshops:
Hands-on, live sessions that guide your team through integrating AI into their workflows and content creation, building practical skills that matter now.
AI Strategy & Implementation Sessions (1:1):
Personalized, one-on-one sessions to uncover your biggest AI opportunities, whether it’s automation, adoption, or designing a tailored roadmap for success.
AI Advisory & Consulting (Retainer or Project-Based):
Ongoing support to embed AI tools into your core processes, automate key workflows and future-proof your communications strategy with measurable impact.

Before you go: I need your brain for two minutes

If you're a consultant, coach, or solo operator, you already know the deal. You don't just do the work. You also have to make everything that sells it: the website, the PDFs, the pitch decks, the landing pages. And you don't always have the budget for a good designer.

I'm building a live, hands-on workshop to fix exactly that, using AI to turn out branded assets that actually look like you, and to leave you with a reusable system instead of a pile of one-offs. It's called The Claude Design System.

Before I finalize the curriculum, I want to build it around your real pain points, not my assumptions. So I made a short survey. Two minutes, mostly multiple choice. Please fill it out, and you'll help shape the workshop, plus get first access to the founding cohort.

👉🏾 Take the 2-minute survey: https://maven.com/forms/0863e6

Thanks for helping me build this!