Scrum Is Dead — But What
Actually Killed It?
Amazon, Google, Netflix are abandoning Scrum. The industry celebrates the framework's demise. But nobody is asking the real question.
It's official: Scrum is dead. At least if you believe the headlines.
Amazon dropped it. Google engineers call it "a religion that produces reports about future greatness instead of great software." Netflix and Microsoft are experimenting with Kanban, OKRs, and autonomous team structures.
The indictment reads convincingly:
- Meeting overhead: Dailies, sprint planning, retros, grooming — a full workday per week consumed by rituals instead of code. (If you've ever witnessed a dysfunctional standup, you know exactly what I mean.)
- Corrupted metrics: Story points, invented as planning aids, mutated into performance weapons. Teams inflate points to look productive.
- Scaling collapse: What works for seven people collapses at fifty teams. Sprint synchronization becomes a full-time job.
- Innovation-hostile: Rigid two-week cycles stifle spontaneous creativity. Gmail was born in Google's 20% time. Not in Sprint 47.
These are real problems. But they only tell half the story.
The Cause of Death Nobody Mentions
Scrum didn't die from its rituals. It died because organizations treated it as a Tool-Set — a package of ceremonies, roles, and artifacts you "implement" like new software.
Set up the Jira board. Hire a Scrum Master. Configure the sprint cadence. Done. We're agile now.
Except: agility was never a Tool-Set. The four values of the Agile Manifesto — individuals over processes, working software over documentation, collaboration over contracts, responding to change over following a plan — these are Mind-Set statements. Every single one describes an attitude. Not a method.
What most organizations did is the equivalent of buying a child a Stradivarius and wondering why they don't play Beethoven.
The instrument is there (Tool-Set). But where are practice and technique (Skill-Set)? And where is the understanding of what music actually means (Mind-Set)?
Why It Fails: The One-Third Problem
The 3-Sets Method\u2122 identifies three dimensions that must work together for any change to succeed: Tool-Set (tools, structures, processes), Skill-Set (abilities, techniques, knowledge), and Mind-Set (beliefs, assumptions, mental models).
Your outcome is capped by your weakest set:
The weakest set caps the outcome
Liebig's Law of the Minimum as a direction-finder: Tool 5, Skill 5, Mind 2 (on a 1-5 scale) → outcome capped at level 2. Brilliance in the strongest set does not raise the cap. Heuristic, not a formula.
This is exactly what happens in most Scrum implementations. I call it the Cargo Cult of Agility:
Diagnosis: Cargo-Cult Scrum
Tool-Set: complete
Jira, Confluence, sprint board, Definition of Done, burndown charts — all there, all configured.
Skill-Set: superficial
Teams know Scrum theory. The practice of self-organization, empirical work, and genuine feedback? Never trained.
Mind-Set: untouched
Control over trust. Failure as defeat. Velocity as KPI instead of planning aid. The culture hasn't moved an inch.
The result? Predictable. Dailies become status reports for management. Retrospectives can't be honest because mistakes are punished. Velocity gets gamed because it's weaponized as a performance measure.
The rituals run. But they're hollow. Scrum as mechanical theater.
What Big Tech Is Actually Doing (And Why It Won't Work for You)
The alternatives that Amazon, Spotify, and others use are revealing — but not for the reasons the LinkedIn headlines suggest.
Shape Up (Basecamp): Six weeks instead of two. Sounds like a Tool-Set swap. It's not. The core is problem shaping before execution — ensuring the team understands the problem before building solutions. That's a Mind-Set intervention.
Spotify Model: Squads choose their own process. This only works in a culture that lives trust and ownership — not one that prints it on posters. The irony: Spotify itself has since moved away from the "Spotify Model." Their culture matured — and their process evolved with it.
OKRs (Google): Outcomes over output. Sounds great. But OKRs in a control-oriented culture become performance theater just like story points. The tool changes. The dysfunction stays.
None of these alternatives are pure Tool-Sets. They all presuppose a certain Mind-Set. The difference: the companies where they work have already built that Mind-Set — often through years of cultural work.
They're not switching frameworks. They have a culture where different frameworks can work. That's the only difference that matters.
Five Ways to Fail at Agile Transformation
- Cargo-Cult Agility. Tool-Set without Mind-Set. The rituals run, the culture stays command-and-control. The most common case.
- Island Agility. Agile teams in a non-agile organization. Teams are frustrated because their environment kills every change.
- Speed Agility. Confusing agility with acceleration. "We're agile" means "we deliver faster" — not "we learn faster." A dangerous mistake.
- Methods Dogma. One framework for everything. SAFe for the startup, Scrum for operations, LeSS for research — all treated the same, none fitting.
- Bottom-Up Without Top-Down. Teams get agile training, management doesn't. Result: agile teams suffocating in non-agile structures.
The Way Out: Diagnosis Over Framework-Hopping
The solution isn't replacing Scrum with the next framework. That's the same mistake in new packaging — a Tool-Set swap that ignores the actual problem.
The solution is an honest diagnosis: Which set is the real bottleneck?
In my experience working with organizations across 20+ years in IT, the bottleneck is almost always the Mind-Set — in roughly eight out of ten cases. Organizations have the tools. They have (at least basic) skills. What they lack:
Psychological Safety
The belief that mistakes won't be punished. Without it, every retrospective is a farce.
Real Autonomy
Not "you may decide" with the fine print "as long as we agree."
Outcome Thinking
Measuring success by what reaches the customer — not by the speed of internal machinery.
Double-Loop Learning
Not just "are we doing it right?" — but "are we doing the right thing?"
Out of the Scrum Graveyard
If your organization is stuck in dysfunctional Scrum, the first step isn't a framework switch. The first step is a diagnosis with the 3-Sets Inventory (3SI): 24 questions, five minutes, three dimensions.
The result doesn't say "Scrum is to blame" or "you need SAFe." It shows which third you're systematically neglecting. And with that, where intervention actually needs to start.
The Conclusion Nobody Wants to Hear
Scrum isn't dead. Scrum was never truly alive in most organizations.
What died is the illusion that you can implement agility like an ERP system — with a project plan, a rollout date, and training for everyone.
The companies burying Scrum are burying their own failed attempt to treat a cultural transformation as a Tool-Set upgrade.
And if they introduce Shape Up, Kanban, or OKRs with the same attitude, they'll write a new article in three years:
"Shape Up Is Dead — Why Big Tech Is Ditching It."
The problem was never the framework. The problem is the third you don't want to see.
The good news: once you've correctly identified which third is lagging, the path forward becomes clear — and often simpler than another framework migration.
Which third is holding your organization back?
The 3-Sets Inventory shows in five minutes where the bottleneck is — and what you should change first.
Start diagnosis