An AI launch operating system that learns the team it works for.
Five agents that produce the launch. Four agents that learn your internal teammates. One agent that argues with you every Friday. Built around one belief: the ceiling of any campaign is the lowest level of cross-functional alignment behind it.
Why this exists.
Most AI marketing tools today are content generators. They produce more content, faster. They do not change the fundamental problem: a great campaign that misaligns the sales lead is a great campaign that does not ship.
Cadence is built around the people problem, not the content problem. It treats internal teammates as customers — profiles their communication style, decision posture, and risk tolerance — and tailors every output to land with each one specifically.
If your launches are bottlenecked on writing, you don’t need Cadence. If your launches are bottlenecked on getting eight humans to agree on the same thing in the same week, this is the framework for you.
Three layers.
Ten agents. Sequential handoffs. One agent runs Opus because taste matters there more than throughput.
Five agents produce the launch.
Sequential handoffs. Each agent reads BRAND.md, SOUL.md, and the prior agent’s structured output. Every agent flags claims built on digital research alone as unverified by default.
Four agents learn your team.
The differentiator. People Profile runs every Monday. Deliverable Tailor reshapes every Layer 1 output for each teammate. Onboarding Bridge integrates new hires across a 60-day learning protocol — honest about uncertainty until the data earns confidence.
One agent argues with you.
Runs every Friday at 3:00pm. Reads everything the system produced this week. Sends one provocation. Permitted to be uncomfortable. Required to be honest. Silence is a valid output. Filler is not.
Three principles enforced by hooks.
Cadence will reject outputs that violate these. Not as a warning. As a refusal.
The customer is the subject. Never the company.
“We’re proud to” formations are auto-rejected. Every sentence has the customer or the customer’s moment as the subject. The brand is what appears between the lines, not what the lines are about.
Internal teammates are customers, not recipients.
Every output is tailored. There is no “broadcast to the team.” Five teammates get five versions of the same source artifact — same data, different shapes.
Digital research is unverified by default.
Any campaign built purely on AI-aggregated or scraped data is flagged as needing human verification before it ships. The system never pretends it knows what it has not seen in person.
The Jake Test.
Cadence is launching FORGE — a fictional non-alcoholic craft brew for sport bike riders — as its working case study.
Jake is the VP of Sales: twenty-year rider, two track days a year, owns an Aprilia Tuono named Old Glory. He will reject anything that sounds like an agency wrote it about riders. Same launch brief, two ways:
“We’re proud to announce the launch of FORGE, a non-alcoholic craft beverage designed for the active motorcycle lifestyle. FORGE leverages the growing demand for premium, mindful consumption to deliver a category-defining experience for performance-oriented consumers.”
Triggers Jake’s auto-reject in four words. “We’re proud to” makes the company the subject. “Lifestyle” is the word he left a previous role over. “Leverages, mindful, premium, performance-oriented consumers” are exactly the language every dealer would mock.
“Six dealers in NorCal will stock FORGE in the cooler within 60 days. The pitch: first NA beer made for the ride-home moment — track day Saturdays, canyon Sunday afternoons, the parking lot at Laguna. Not Athletic. Not Liquid Death. Not Road King. Riders who already buy three flavors of energy drink for the cooler in the truck. Confidence range: 4–6 dealers say yes in the first 60 days. Back-out: if September lands under 4, pull dealer placement and re-route to event sampling at Buttonwillow. Person to start with: Sergio at Aprilia of San Jose.”
Same source brief. Same data. Reshaped: a real dealer name, a real venue, the competitive set Jake knows, his vocabulary (“cooler in the truck”), a confidence range, a back-out plan, a real first call.
The repo.
Cadence is a forkable framework. Clone it. Replace the brand. Replace the people. Run your own launches. Fork time: 4–6 hours of focused work.
This week’s provocation.
Every Friday at 3:00pm, Cadence’s Status Quo Breaker reads what the system produced this week and sends Garth one provocation. The agent is allowed to argue. It is required to be honest. Silence is a valid output.
Garth — Cadence’s launch brief for FORGE is built around two personas. Diego was built from in-person observation at one dealer in San Jose. Marcus was built entirely from Reddit and YouTube signal. The brief weights them equally in the channel mix.
You wrote in SOUL.md that you cannot research a customer from behind a laptop. Half of FORGE’s launch is being researched from behind a laptop. The framework you built is producing a campaign that violates the framework’s own non-negotiable.
Either Marcus needs in-person validation before Aug 15, or the channel mix needs to be rebalanced so Diego carries the launch alone. Which is it?
About the human in the system.
I’m Garth Beyer. I’m a product marketing manager with a demand-gen lifecycle lean, building toward a CMO seat. Cadence is the framework I built while thinking through how AI changes the shape of senior PMM work.
Most AI marketing tools optimize for content velocity. I’m more interested in what AI does to the cross-functional alignment problem — the actual reason most launches underperform. Cadence is what came out of that thinking. It is opinionated. It will reject your draft. It will argue with you on Friday. That’s the point.
If you’re hiring for a senior PMM or building an AI-native marketing function and want to talk — the email below is the right place.
Cadence runs on a private Friday cadence by default. Want to fork it for your org? Download the .zip above. Pages don’t lie about what they are.
