Thank you, Scrum

Published:

Scrum stands as a rare and transformative management innovation in a field where technical progress has accelerated but management practices have stagnated.

Imagine a developer from 1995 has time-traveled to 2025: the development techniques aren’t just faster, they are different paradigms (open source, AI as API, infrastructure as code, CI/CD, multi-threading and concurrency are ubiquitous).

Then imagine a manager from 1995 has time-traveled to 2025: they’d easily recognize the most prevalent patterns: Gantt charts, project plans, performance reviews, 1-on-1s, management by objective, same budgeting, same focus on utilization of resources. They’d be confronted with new tech and software, sure (email is ubiquitous, slack channels, etc), but the only new management patterns they’d experience would be the rather ubiquitous use of Scrum (or parts of Scrum).

I See A Disparity Between Technical and Managerial Evolution

Software Development Technique Has Radically Changed in the Past 30 Years (1995 → 2025)

19952025
Open source was fringe.It’s now the backbone of nearly every system.
Creative Commons didn’t exist.Its licensing model reshaped IP norms.
Multi-threading and concurrency was rare, expensive, and mostly academic.Multi-core processors and concurrency are standard.
Monolithic architectures dominated.Microservices, monorepos, distributable packages prevail.
Development tools were expensive.Today’s tools are free and abundant.
SaaS didn’t exist.Most software is now subscription-based.
CI/CD was unheard of.Daily deployments with zero-downtime are routine.
Virtualization was rare.Now it's ubiquitous.
The XP book hadn’t been published.DevOps (a descendant of XP) is mainstream.
AI was science fiction.AI and ML models run in our pockets.

Summary: Tooling, scale, and delivery patterns have all evolved beyond recognition. What is normal today would appear as alien technology to a developer in 1995.

But Management Technique Hasn’t Changed

Most organizations are still managed like they were a century ago.

  • Carrot-and-stick reward schemes.
  • Command and control structures.
  • Taylorism rebranded as OKRs and KPIs. (Management-by-Objective with new acronyms.)
  • Rigid roles, top-down planning, performance reviews.
  • Division of labour and strict separation of roles and responsibilities.

Conclusion: We’re managing 21st-century software work with 20th-century playbooks.

With One Exception: Scrum

Hate it or love it, Scrum is here to stay. I think that is a net positive.

When Steve Denning wrote in Forbes Magazine that Scrum Is A Major Management Discovery, I thought he was going a little too far. I sat beside him at a conference in Orlando (he likely wouldn’t remember) and mentioned the article — I had read it and thought he was exaggerating. It was a well-written article, interesting, but I had always considered Scrum to be “radical common sense” (a phrase I used a lot in those days). How could something I considered “common sense” be so revered. Denning even argued the inventors of Scrum may be worth of a Nobel Prize (“If there was a Nobel Prize for management”, he noted.) 10+ years has passed since that article and I’m more inclined to agree with Denning.

To a time-traveller from a century ago, most workplace “management” practices would appear unchanged, except for Scrum (or parts of Scrum) that have taken hold (almost worldwide) as the standout management innovation.

Scrum Didn’t Come From Academia. It Came From the Trenches.

It is worth noting that Scrum was devised by software practitioners, not theorists. It is grounded in empirical process control theory rather than dogmatic, outdated assumptions about human nature and motivation that emphasize processes and tools over individuals and interactions. It is grounded in pragmatic patterns and has essentially entrench iterative & incremental development patterns in mainstream business.

Scrum challenged conventional management dogma.

Scrum’s Paradigm Shifts

Conventional ThinkingScrum’s Alternative
Supervision is necessary. Visibility is control.Self-management empowers teams. Transparency improves shared understanding and accelerates decision-making.
Large batch processing. Sequential/gated Project Management.Small batch processing. Iterative & Incremental Delivery.
Division of labour.Integration and fluidity of skills.

The Ripple Effect: A New Way of Thinking Took Hold

Scrum Opened the Door for Adjacent and Subsequent Innovations

While the following are not direct descendants, Scrum’s was like an icebreaker: figuratively and literally. Attending a Scrum training course, for example, caused many to challenge long-held assumptions about managing product development. Then, having disconfirmed what they previously believed, many were open to explore other empirical/data-driven strategies.

  • Lean Startup: build-measure-learn.
  • UX: user-centered iteration.
  • DevOps: small-batch flow, feedback, automation, continuous delivery.
  • Kanban: throughput metrics, flow optimization, demand:capacity balance.

Scrum Served As An Example Of

  • Empirical process control theory
  • Teams as the performance unit
  • Product development vs. Project management

Some got it, others misunderstood — either way, Scrum shifted the landscape.

Like It Or Not…

  • Millions of developers now take for granted:
    • Iterative & incremental delivery (IID).
    • Small-batch, continuous delivery.

In the domain of technology, eXtreme Programming and technological advancements are due credit.

In the domain of management, Scrum is due credit. “Agile” gets a lot of credit, but I disagree — perhaps that’s for a future essay.

And while Scrum has many haters, ask any elder developer — before Scrum, most enterprises were much worse than today.

Thank you, Scrum.

I’m glad some of the assumptions that guided management in 1995 have been invalidated, replaced with Iterative & Incremental Development patterns. This is largely due to Scrum having pioneered new management patterns that disrupt previous conventions.

And there’s still much to learn and I’ll be disappointed if management in 2055 (30 years from now) has again stagnated.