Keep the Flame Alive: The High-Speed-Train Comms Cadence
Weekly bite-sized, honest updates and 3-minute customer snippets build momentum no one wants to step in front of
- Difficulty
- Easy
- Time to result
- ~ongoing to results
- Steps
- 3
- Confidence
- 90%
When a new bet has nothing shippable to show for months (the 'ugly baby' phase), internal communications is everything. Broadcast honest, bite-sized progress every single week — early on the momentum of the build, later 3-minute customer video snippets — so hundreds of colleagues perceive a fast-moving train. People don't want to get in front of a high-speed train, so consistent visible velocity buys you the time and protection to survive the phase where you have no polished outcome yet.
Origin
Tanguy Crusson's practice on Jira Product Discovery, using Atlassian's internal 'Atlas' tool for weekly tweet-sized project updates; echoes Mahika Kapoor's (Figma) 'keep the flame alive and help it spread.'
Core principles
- 01When there's nothing to show, internal comms is everything
- 02Be brutally honest — don't inflate progress; report what you're testing and learning
- 03People consume bite-sized weekly content; they won't read a 30-minute report or absorb a three-month info dump
- 04Momentum is perceived — a high-speed train deters interference
- 05Customer snippets hit both heart and mind better than research reports
How to run it
- 1
Commit to a weekly public update
Post a tweet-sized update every week to a subscribable internal channel about where you are, what you're learning, and how fast. Aim for honesty over spin.
Pro tip Crusson's JPD updates drew hundreds of internal subscribers who followed along weekly.
Watch out Don't try to make it sound bigger or more successful than it is — honesty every step is what earns durable trust.
- 2
Early on, show the build momentum
Before customers are involved, run a weekly demo of the product and everything built, emphasizing forward motion.
Pro tip Save a feature or two in reserve so you always have something to show — e.g. before a vacation week — to keep the train visibly moving.
- 3
Later, lead with 3-minute customer snippets
Once customers are engaged, replace build-demos with short snippets of customer conversations — problem, then how you solve it, in their words — and share them widely on Slack.
Pro tip No one watches a 30-minute research report; everyone will watch a 3-minute snippet of four customers talking about a problem.
In the wild
Crusson used Atlassian's Atlas tool for weekly tweet-sized updates and weekly demos, saving features to sustain visible momentum. He had dogfooding within two months, the first lighthouse user at five months, ~six months of alpha, then nearly a year of beta — and at every step people could see the team going somewhere, so they judged the team's trajectory rather than a premature outcome.
→ The steady cadence bought the time to get through the no-visible-outcome phase; JPD reached GA and 8,000 customers within a year of launch.
Common mistakes
Going dark then resurfacing with a big reveal
Coming back three or four months later with 'here's what we have to show' fails — people can't absorb it. They readily consume small weekly increments but struggle with a large infrequent dump.
Is it for you?
Best for
A 0-to-1 lead who needs to protect a bet through months of no shippable outcome inside a larger org
Not ideal for
Mature products where a stable weekly hype cadence would be noise rather than protection
From the transcript
“basically internal coms is everything”
“those kinds things are what give people a sense of velocity and speed and no one wants to”
“no one is going to read a research report that takes 30 minutes to read everyone is happy to watch a three minute snippet”
“I was even saving features for one week to the next just to when when I knew we were going to be on vacation”
“weekly you send a tweet sized update about your project”
From the episode
Hard-won lessons building 0 to 1 inside Atlassian
Tanguy Crusson (Head of Jira Product Discovery)