Ryan Singer (creator of “Shape Up,” early employee at 37sign30 March 2025

A better way to plan, build, and ship products

6Frameworks

14Insights

Episode overview

Ryan Singer explains Shape Up, the product-development method that grew from 37signals' effort to preserve a fast, founder-led way of working as the company expanded. The approach works backward from a fixed appetite, shapes a concrete solution collaboratively with product, design, and engineering, and gives builders responsibility for implementation. Singer also explains how real-world organizations can pilot the method, why incomplete shaping causes projects to overrun, and how product managers can move upstream toward problem framing and product strategy.

Key ideas

Set an appetite—the maximum time worth spending—before shaping the scope, with six weeks serving as an upper limit rather than a mandatory cycle length.
Do not start a project until the team can see the end from the beginning and identify the main moving pieces, dependencies, and technical risks.
Shape solutions collaboratively with product, design, and an experienced engineer who understands the existing system and can expose hidden complexity.
The output of shaping should be concrete enough that builders know what to build, but low-fidelity enough to preserve room for detailed design and implementation decisions.
Give builders the whole shaped idea and let them create their own tasks; a kickoff exercise can summarize implementation in nine or fewer major scopes.
If a project is not on track, move it back from build mode into shaping mode instead of blindly extending it, arbitrarily cutting essential value, or continuing unclear investment.
Adopt Shape Up through a meaningful pilot project whose problem, appetite, team, and shaping window fit together, rather than copying only the six-week cycle.
In a mature Shape Up team, product managers spend less time shepherding execution and more time framing valuable customer and business problems upstream.

Transcript available · source text is retained privately and is not published

Frameworks in this episode

Productivity3 steps

Fixed Time, Variable Scope (Appetite-Based Planning)

Set the max time you'll spend first, then design a solution that fits it — instead of estimating a concept.

Read →

Strategy3 steps

Framing (Setting the Boundaries)

Narrow a fuzzy request down to the specific problem worth solving before you design any solution.

Read →

Innovation5 steps

Live Shaping Session

Put product, a senior engineer, and a designer in a room to sketch and break ideas until one is clearly buildable in the appetite.

Read →

Productivity3 steps

The Circuit Breaker (Back to Shaping)

When a project overruns its appetite, don't extend it or gut the agreed value — pull it back into shaping mode.

Read →

Leadership4 steps

The Detail Dial

Calibrate how prescriptive a shaped plan is to the builder's seniority — more detail lifts juniors, less frees seniors.

Read →

Productivity3 steps

The Nine-Scopes Kickoff

At kickoff, have the builders translate the shaped idea into nine-or-fewer implementation chunks to test scope and build clarity.

Read →

People & resources mentioned

Attributed to the moment in the episode. Timestamps are approximate.

People · 8

Ryan SingerMentionsRyan was one of the first few hires at 37 signals
“Ryan was one of the first few hires at 37 signals”
▶ Watch at ~0:30approx.
Lenny RachitskyMentionstoday we've got another very special compilation episode something I've been pulling on more and more with the podcast and the newsletter
“today my guest is Ryan Singer”
▶ Watch at ~0:30approx.
Des TraynorMentionsIntercom co-founder who alerted McCabe to the AI team's early product and is writing a book about AI disruption.
“a big thank you to Des trainer Bob mea and Chris spec for suggesting questions and topics”
▶ Watch at ~1:30approx.
Chris SpiekMentionsa big thank you to Des trainer Bob mea and Chris spec for suggesting questions and topics
“a big thank you to Des trainer Bob mea and Chris spec for suggesting questions and topics”
▶ Watch at ~1:30approx.
Jason FriedMentionstoday my guest is Jason freed Jason is the co-founder and CEO of 37 signals
“it was Jason's idea to even write the book”
▶ Watch at ~8:00approx.
David Heinemeier HanssonMentionsDavid hires programmers and different people in the company hire different people
“this is dhh uh he was working part-time when when he started 37”
▶ Watch at ~11:30approx.
Bob MoestaMentionsBob is the co-creator of the job is to beone framework and worked alongside Klay Christensen for many years
“Bob mea who's been on your show a couple times”
▶ Watch at ~25:30approx.
Clayton ChristensenMentionsManagement thinker admired by Eric Ries whose endorsement of Ries's work was personally important to him.
“that's the one that Clayton Chris uh Klay Christensen wrote”
▶ Watch at ~1:41:00approx.

Resources · 22

37signalsMentionscompany · Jason Fried and co-founders
“Ryan was one of the first few hires at 37 signals”
▶ Watch at ~0:30approx.link ↗
Shape UpCoinedbook · Ryan Singer
“he wrote a book called shape up which shares a very different approach to building software”
▶ Watch at ~1:00approx.link ↗
BasecampMentionssoftware · 37signals
“through his experience of building Base Cam and a 17 years of building products at 37 signals”
▶ Watch at ~1:00approx.
Lenny's NewsletterMentionsnewsletter · Lenny Rachitsky
“if you become a paid subscriber of my newsletter you get an entire year free”
▶ Watch at ~1:30approx.
NotionRecommendssoftware · Notion Labs
“you get an entire year free of perplexity Pro notion superhuman linear and granola”
▶ Watch at ~2:00approx.
GranolaRecommendssoftware · Granola
“you get an entire year free of perplexity Pro notion superhuman linear and granola”
▶ Watch at ~2:00approx.
LinearRecommendssoftware · Linear
“you get an entire year free of perplexity Pro notion superhuman linear and granola”
▶ Watch at ~2:00approx.
SuperhumanRecommendssoftware · Superhuman
“you get an entire year free of perplexity Pro notion superhuman linear and granola”
▶ Watch at ~2:00approx.
Perplexity ProRecommendssoftware · Perplexity AI
“you get an entire year free of perplexity Pro notion superhuman linear and granola”
▶ Watch at ~2:00approx.
WarrantMentionssoftware · Warrant
“workos also recently acquired warrant the fine grain authorization service”
▶ Watch at ~2:30approx.
ZanzibarMentionssoftware · Google
“warrant's product is based on a groundbreaking authorization system called Zanzibar which was originally designed for Google”
▶ Watch at ~2:30approx.
WorkOSRecommendscompany · WorkOS, Inc.
“you should consider work OS it's a drop-in replacement for op zero”
▶ Watch at ~3:00approx.
MergeMentionscompany · Merge
“the folks at merge are obsessed with Integrations their single API helps SAS companies launch over 200 product Integrations”
▶ Watch at ~3:30approx.
Shaping in Real LifeCoinedcourse · Ryan Singer
“this led to this uh course that we made which is called shaping in real life”
▶ Watch at ~9:30approx.
RubyUsessoftware · Yukihiro Matsumoto
“he told Jason I want to try building this in Ruby”
▶ Watch at ~11:30approx.
Ruby on RailsCoinedsoftware · David Heinemeier Hansson
“the framework Ruby on Rails he he ended up releasing that after base camp was standing”
▶ Watch at ~12:00approx.
FigmaMentionssoftware · Figma
“it wasn't making a beautiful figma file to represent a concept that could maybe be a feature”
▶ Watch at ~15:00approx.
Google CalendarMentionssoftware · Google
“we're not going to just build calendar which is like Google Calendar”
▶ Watch at ~42:00approx.
Airtable Product CentralMentionssoftware · Airtable
“this episode is brought to you by air table product Central”
▶ Watch at ~46:00approx.
Demand-Side Sales 101Recommendsbook · Bob Moesta and Greg Engle
“the one that I recommend the most is a demand side sales one 101”
▶ Watch at ~1:40:00approx.
Competing Against LuckRecommendsbook · Clayton M. Christensen, Taddy Hall, Karen Dillon, and David S. Duncan
“competing against luck is a really good intro”
▶ Watch at ~1:41:00approx.
Lenny's PodcastCoinedpodcast · Lenny Rachitsky
“you can find all past episodes or learn more about the show at Lenny podcast.com”
▶ Watch at ~1:45:00approx.

Spot an error or want something removed? Request a correction or removal.

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Myth Buster· 2

Myth Buster32:00

Why You Can't Just 'Cut Scope' When Time Runs Out

A seductive misreading of Shape Up says: fix the time, and if you run out, just cut scope like Parkinson's Law. Singer says that works for a low-stakes landing page but not for real product work where functionality has to actually exist. If the work wasn't shaped, cutting the agreed-upon value at the deadline kills morale and leaves the team with a debt feeling going into the next cycle. The real answer is doing the shaping work up front.

Parkinson's Law logic (cut scope at the deadline) only fits trivial work like a landing page
Real product functionality can't simply be trimmed away when time runs out
Cutting agreed value at the deadline destroys morale and creates a 'debt feeling'
The fix is upfront shaping, not last-minute scope-cutting

“but when it comes to real product work you know where there's some functionality that we have to figure out how to make it exist…”
Ryan Singer · 32:30

#shape-up#scope#parkinsons-law#shaping

Myth Buster47:00

The Real Failure Mode Is Too Little Detail, Not Too Much

PMs fear that a detailed shaped pitch turns engineers into code monkeys. Singer says the dominant real-world failure is the opposite: not enough detail, with engineers running back to product saying 'I'm not getting enough from you.' The amount of detail that helps is a dial you turn based on who's on the team — more guidance makes a junior engineer far more successful (and stops them from hiding that they're lost), while stellar senior talent can be given less.

The most common failure is too little detail, not over-specification
Engineers frequently complain they aren't getting enough from product
Detail is a dial tuned to the seniority of the person building
Junior engineers hide when they're lost; more guidance sets them up to succeed and learn

“the dominant failure case that I see in the real world is always again and again not enough detail”
Ryan Singer · 47:30

“the amount of detail that the team is going to feel helps them is a dial that we can turn that depends on who on…”
Ryan Singer · 48:00

#shape-up#shaping#product-management#seniority

Hot Take· 2

Hot Take1:18:30

A Feature Factory Is Actually Healthy — Just Feed It Better Input

Singer reframes the popular 'feature factory' critique. If your org is reliably cranking out features, you're probably quite healthy — you just need to feed a different input (real problem framing and demand) into the front of the factory. The far more widespread and painful struggle he sees isn't over-shipping; it's teams that can't ship predictably at all, where work drags, the end is invisible, and people burn out.

A team that reliably ships features is healthy; fix the input, not the output engine
The feature-factory critique is really about not negotiating value and outcomes upfront
The bigger real-world problem is unpredictable, non-repeatable shipping
Dragging work with no visible end is what actually burns teams out

“if you have a feature Factory meaning you're continually cranking out features you're probably quite healthy all you need to do is feed a different…”
Ryan Singer · 1:19:00

“what most teams are struggling with is that they can't they can't they don't have predictable repeatable shipping of things”
Ryan Singer · 1:19:00

#shape-up#feature-factory#shipping#framing

Hot Take1:28:30

Basecamp Was Uniquely Set Up — Don't Blindly Copy It

Singer only realized how unusual Basecamp was after leaving. Every designer there actually coded — running the app locally and rendering views themselves — so the painful wall between design and engineering simply didn't exist. There was also no sales or marketing org creating contention for engineering time; the founders did all of that, and stayed close to problem definition. Copying Basecamp's advice without those conditions is why Shape Up often stumbles elsewhere.

Every Basecamp designer coded, not just HTML — they ran the app and rendered their own views
The disappointing 'engineer says no to the Figma file' moment didn't exist there
No sales or marketing meant no contention for engineering time
Founders stayed close to problem definition, keeping constant clarity on what and why

“every designer codes imagine if every designer codes and not and I don't just mean html I mean like running the app locally”
Ryan Singer · 1:29:00

“imagine you have no sales or you have no marketing that all of sale selling and marketing is happening by the Unicorn Founders so it…”
Ryan Singer · 1:30:30

#basecamp#shape-up#team-structure#context

Explainer· 2

Explainer35:30

Shape Up Works at Big Companies — and Not the Whole Company Has To

Singer counters the assumption that Shape Up only fits tiny Basecamp-like teams. He cites a pharmaceutical company with thousands of people where a few teams run it successfully. The unlock is good engineering leadership that untangles dependencies so one system can be worked on independently, plus capacity management to protect a team from being pulled onto other work for a set number of weeks.

A thousands-person pharma/clinical-trials company runs Shape Up on a few key teams
Dependency hell is not inevitable — good engineering leadership untangles it
A senior engineer must make architectural choices and back-stop the team from being pulled away
Only part of a large company needs to adopt it, which many find freeing

“so many people are used to it and they think that it's just how it is but it's actually not it is possible for engineering…”
Ryan Singer · 36:30

#shape-up#scaling#engineering-leadership#dependencies

Explainer40:30

The Home Renovation Analogy: Check for Electricity in the Wall

Singer's signature analogy for why beautiful designs mislead. You can have a perfect rendering of a bedroom with wall-mounted lamps, but if you never checked whether there's electricity in that wall, the real cost and time change drastically once you have to rip the walls open. A high-fidelity Figma file hides the same unknowns — the engineering team has to put on x-ray goggles to figure out the logic and flows underneath the UI.

A perfect rendering means nothing if you haven't checked what's behind the wall
Unseen wiring (or backend logic) drastically changes cost and time
So much of a feature's real work is invisible on the surface of the UI
Engineers must 'x-ray' a design to understand flows, cases, and behind-the-scenes logic

“you can like have the perfect rendering and the perfect lamp and the perfect color but if you haven't checked if there's electricity in that…”
Ryan Singer · 41:00

#shape-up#shaping#design#analogy

Story· 4

Story11:00

The 10-Hours-a-Week Constraint That Seeded Shape Up

Singer traces Shape Up's origins to the first version of Basecamp, when DHH (David) was only programming about 10 hours a week while building the app. That scarcity of engineering time, combined with Jason Fried's relentless push for visible forward movement, forced the team to be ruthless about not wasting a single build cycle. The pressure to use David's limited time well is what pushed them toward figuring out ideas sharply before committing to build them.

David (DHH) was programming only ~10 hours a week on Basecamp V1
That constraint created intense pressure to use engineering time efficiently and never build the wrong thing
Jason Fried's constant demand for movement added cultural urgency even with no outside pressure
Avoiding wasteful build-and-throw-away cycles became the founding motivation

“even when we were building V1 David wasn't actually full-time as the as our only technical person he was programming 10 hours a week so…”
Ryan Singer · 11:00

#shape-up#basecamp#engineering-time#origin-story

Story17:30

The Project That Forced Basecamp to Formalize Shape Up

For roughly the first decade, the natural startup way of working spread organically as Basecamp hired slowly. Then one project broke the pattern: a review session revealed lots of open questions, no quick answers, and the dawning realization that not only would it not ship, they couldn't even see its end. That moment convinced Singer the organic method wouldn't scale forever, and he took on systematizing it into what became Shape Up.

For ~10 years the way of working spread organically because founders hired deliberately slowly
One project's review session surfaced many open questions and slow answers
The team realized it wouldn't ship and they couldn't see the end of it
That failure triggered Singer to formalize and systematize the method

“what we're starting to realize is like oh this is not only is this not going to ship but we can't even see the end…”
Ryan Singer · 18:00

#shape-up#basecamp#scaling#origin-story

Story50:30

The Fintech Onboarding Step That Was Secretly Three Steps

A fintech team wanted to remove a high-drop-off onboarding step by piping customer data from partner banks — a clear conversion win, on paper. What they missed: in the code, that single step was actually three different branches depending on which bank the customer was integrated with. Discovering that in week four is a disaster; discovering it in the shaping room, before greenlighting, turns it into a productive tradeoff conversation. Singer frames the shaping engineer as the 'grumpy old plumber' who opens the walls before quoting.

A fintech team planned to cut an onboarding step by importing data from partner banks
The one step was actually three code branches depending on the customer's bank
Finding this in week four (after resourcing) is painful; finding it in shaping is productive
The shaping engineer is like a plumber who opens the walls to inspect the pipes before quoting

“the thing that they didn't look at was if you go into the code on that step of the onboarding it's not actually one step…”
Ryan Singer · 51:00

“the grumpy old plumber who's seen everything and he insists on opening up the walls and looking at the pipes before he'll give you a…”
Ryan Singer · 52:30

#shape-up#shaping#risk#fintech

Story1:33:30

Why Shape Up 'Didn't Work' — No Engineer in the Room

After the book, Singer heard from teams whose projects kept running over. When he asked to see their shaping work, they'd show him a PRD or a stack of Figma files — nothing like the book. The pattern: people in existing roles kept producing their usual artifacts, and there was no engineer in the shaping picture. His breakthrough came on a consulting project where they brought the best-suited engineer over into product for the shaping session, and suddenly he was 'in the world' he knew again.

Teams reported projects constantly running past the cycle
Their 'shaping work' was actually PRDs and Figma files, unlike the book
The root cause: no engineer was present in the shaping
The fix that unlocked it: bring the right engineer into product for the shaping session

“it wasn't that this doesn't work I was just in a like a foreign country you know it was like um we tried it and…”
Ryan Singer · 1:33:30

“can you show me your shaping work and then they would show me a PRD I'd be like like well that's not that doesn't look…”
Ryan Singer · 1:34:00

#shape-up#shaping#consulting#engineering

Tool· 2

Tool1:06:00

The Nine-Boxes Kickoff Exercise (Rule of 7±2)

For mixed-seniority teams, Singer uses a simple kickoff exercise: take what was shaped and have the builders draw a grid of nine boxes representing the nine major chunks of implementation. It surfaces whether the scope is too big, and creates coaching moments where seniors correct juniors' approaches. If it needs more than ~10 boxes you've fallen into ticket land; nine or fewer maps to the cognitive 'rule of 7±2' — a whole you can hold in your head.

At kickoff, builders translate the shaped work into nine major implementation chunks
Six weeks / nine boxes is roughly four business days per box
More than ~10 boxes means you've dropped into unhelpful ticket land
Nine or fewer maps to the 7±2 cognitive limit — a whole you can hold in your head

“give me nine boxes of the nine major chunks that you think have to get implemented from an implementation standpoint”
Ryan Singer · 1:06:30

“if it's more than 10 then you just get into ticket land of here's a million things I have to do”
Ryan Singer · 1:08:00

#shape-up#kickoff#estimation#tools

Tool1:38:30

Book Pick: 'Demand-Side Sales 101' for the Framing Step

Singer connects shaping's upstream 'framing' step to product strategy and jobs-to-be-done. When projects fail because the very input is unclear — you don't know what's important to customers or where the value is — he reaches for Bob Moesta's work. His top recommendation is 'Demand-Side Sales 101' (a tactical jobs-to-be-done book despite its off-putting title), with Clayton Christensen's 'Competing Against Luck' as a good general intro to the spirit of jobs-to-be-done.

Many shipping failures trace back to an unclear input at the very start (framing)
Framing links Shape Up to product strategy and jobs-to-be-done
Top pick: Bob Moesta's 'Demand-Side Sales 101' — tactical despite the sales-y title
'Competing Against Luck' by Clayton Christensen is a good intro to jobs-to-be-done

“the one that I recommend the most is a demand side sales one 101”
Ryan Singer · 1:40:00

“competing against luck is a really good intro that's the one that Clayton Chris uh Klay Christensen wrote”
Ryan Singer · 1:41:00

#jobs-to-be-done#framing#books#product-strategy

Takeaway· 2

Takeaway31:00

The Circuit Breaker: Pull a Stuck Project Back Into Shaping

Basecamp's original rule was a 'circuit breaker' — if a project isn't on track to finish, cancel it and rethink. Singer admits almost no team has the stomach to just kill projects. The more palatable version: don't keep reinvesting in something you don't understand. Take it out of build mode and bring it back into shaping mode, which may mean different people, a different conversation, and different questions to find what's fuzzy.

Basecamp's circuit breaker: cancel a not-on-track project and rethink it
Almost no team has the stomach to actually cancel
Better move: stop reinvesting in what you don't understand
Pull the project out of build mode and back into shaping mode with fresh people and questions

“the circuit breaker like if a project is not on track to actually finish after the six weeks we're just going to cancel it and…”
Ryan Singer · 31:00

“let's take this out of build mode and bring this back into shaping mode which might mean different people a different conversation asking different questions”
Ryan Singer · 31:30

#shape-up#circuit-breaker#shaping#project-management

Takeaway1:26:00

In Shape Up, the PM Moves Upstream

Singer argues that PMs today often get stuck babysitting the build phase — closer to project management than product management. In teams that hit their Shape Up stride, the PM moves upstream: away from keeping a project out of a bad state, and toward understanding business context, narrowing the problem, and negotiating with leadership about which slice of a problem is actually worth spending weeks on. Lenny notes AI-assisted building may push the PM role in the same upstream direction.

PMs often over-invest in shepherding the build, which is really project management
In mature Shape Up teams the PM shifts upstream to problem definition and framing
The high-leverage PM work is understanding the business, customer, and which problem slice is valuable
AI-assisted building may pull the PM role further upstream toward deciding what to build

“what we see in really In Shape Up teams when they hit their stride is that the PM moves Upstream”
Ryan Singer · 1:26:30

“that really getting deeper understanding of the business and the problem and the customer domain and like what problem is worth solving”
Ryan Singer · 1:27:00

#shape-up#product-management#framing#ai

Key ideas

Frameworks in this episode

Fixed Time, Variable Scope (Appetite-Based Planning)

Framing (Setting the Boundaries)

Live Shaping Session

The Circuit Breaker (Back to Shaping)

The Detail Dial

The Nine-Scopes Kickoff

People & resources mentioned

People · 8

Resources · 22

Insights & moments

⚡Myth Buster· 2

Why You Can't Just 'Cut Scope' When Time Runs Out

The Real Failure Mode Is Too Little Detail, Not Too Much

◆Hot Take· 2

A Feature Factory Is Actually Healthy — Just Feed It Better Input

Basecamp Was Uniquely Set Up — Don't Blindly Copy It

✶Explainer· 2

Shape Up Works at Big Companies — and Not the Whole Company Has To

The Home Renovation Analogy: Check for Electricity in the Wall

❝Story· 4

The 10-Hours-a-Week Constraint That Seeded Shape Up

The Project That Forced Basecamp to Formalize Shape Up

The Fintech Onboarding Step That Was Secretly Three Steps

Why Shape Up 'Didn't Work' — No Engineer in the Room

✦Tool· 2

The Nine-Boxes Kickoff Exercise (Rule of 7±2)

Book Pick: 'Demand-Side Sales 101' for the Framing Step

▲Takeaway· 2

The Circuit Breaker: Pull a Stuck Project Back Into Shaping

In Shape Up, the PM Moves Upstream

Myth Buster· 2

Hot Take· 2

Explainer· 2

Story· 4

Tool· 2

Takeaway· 2