The Low-Maintenance Work-in-Progress Deck
One shared slide deck of screenshots, updated monthly, that kills redundant work and spreads patterns.
- Difficulty
- Starter
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 92%
A ritual Katie Dill has run for over a decade, from Airbnb to Stripe. As companies grow, people lose sight of what's happening elsewhere; the artifacts that exist (PRDs full of jargon and project codenames) do not create clarity. Instead, every designer adds a screenshot or prototype of what they're working on to one shared Google Slides deck on a cadence, and the deck is circulated to designers, PMs, engineering leaders, and company leadership. A picture tells a thousand words; a prototype saves a thousand meetings.
Origin
Katie Dill's own practice, started at Airbnb and carried to Lyft and Stripe — she describes it as something she will take wherever she goes.
Core principles
- 01PRDs full of words and project codenames do not create clarity; images do.
- 02The tool must be low-maintenance or the ritual dies — friction, not value, is what kills it.
- 03Accessibility to non-designers beats fidelity for designers: use the tool everyone already has.
- 04It is not a status check — it is a picture of what's happening.
- 05Showing work in progress feels exposing but prevents worse pain at the end.
How to run it
- 1
Choose the lowest-friction shared surface
Use a plain shared slide deck (Google Slides), not the design tool. Designers would prefer Figma, but the deck is flippable by any function, easy to email, and everyone already knows how to comment on it.
Pro tip Optimize the tool for the least design-native reader in the audience.
Watch out If not everyone in the company lives in your design tool, choosing it excludes exactly the audience the ritual exists for.
- 2
Ask for a screenshot or prototype per project
Each designer adds one slide — a screenshot or prototype of what they are working on. Visual only; no status narrative.
Pro tip Make explicit that it is work in progress and not up for detail critique, so people share early rather than polishing.
Watch out Designers will want to build beautiful mocks for it. That turns the ritual into an arduous task and kills it.
- 3
Set the granularity so the deck stays flippable
Filter by project size — ask for medium and large projects only if the deck balloons. A 200-page deck nobody flips through has failed.
Pro tip Tune granularity to team size; expect to re-tune as the team grows.
- 4
Run it on a cadence and circulate widely
Stripe currently runs it monthly (previously bi-weekly). Send it beyond design — to PMs, engineering leaders, and company leadership — so they see what is being built across the journey.
Pro tip Experiment with cadence: bi-weekly is more fun but tips into arduous; monthly has been holding.
Watch out If it starts feeling like an arduous task, it is not succeeding — reduce the ask, don't push harder.
- 5
Harvest the collisions and the patterns
The payoff is designers spotting each other ('you're working on that surface — so am I, let's talk'), pattern reuse ('that's interesting, we could use this in more places'), and — critically — catching duplicated or colliding projects early.
In the wild
Without shared visibility, teams work tirelessly for long stretches and discover at the end that another team is doing the same project, or that the two are on a collision path.
→ The deck surfaces the overlap early, saving time and producing better work — the stated reason it is worth the discomfort of showing unfinished work.
Because the deck goes to PMs, engineering leaders, and executives, they can see what's happening on the marketing side alongside the product side and understand how the pieces of a user journey fit together.
→ Leadership gains a journey-level picture of what the company is building, without additional meetings.
Common mistakes
Turning it into a status report
The moment it reads as 'show us what you're doing', it becomes surveillance and people optimize the slide instead of the work. It is a picture of projects, not a performance artifact.
Letting perfectionism raise the cost
Designers will polish. If contributing takes real time, participation decays and the deck dies. Explicitly frame it as work in progress, not ready to critique.
Choosing the tool designers prefer
Picking the high-fidelity design tool locks out the non-design audience — the very people whose visibility makes the ritual valuable.
Is it for you?
Best for
Design, product, or engineering leaders of teams past ~30 people where people have lost sight of what other teams are building and duplicate work is appearing.
Not ideal for
Small co-located teams who already see each other's screens daily, or organizations where every function genuinely lives in the same design tool.
From the transcript
“I'm strong believer that a Picture Tells a thousand words and a prototype tells you know saves a thousand meetings”
“having people within the design team share as a screenshot or a prototype of what they are working on in a shared deck”
“one of the key pieces is just like keeping it really low maintenance”
“it is not meant to be a status check we're not asking everyone like show us what you're doing”
From the episode
Building beautiful products with Stripe’s Head of Design
Katie Dill (Stripe, Airbnb, Lyft)