You’ve heard it. Maybe you’ve said it.
“Agile doesn’t work for…”
- Big initiatives
- Regulated products
- Hardware
- Fixed-price contracts
- Safety-critical systems
The problem isn’t that the agile values and principles are wrong. It’s that too often, typical agile practices assume a context that doesn’t match where most people actually work.
Agile approaches do produce great outcomes for customers and create more human-friendly work environments. But when your boss wants a detailed plan and the business needs to know when things will be done, it’s hard to make the case.
As one senior engineering leader told us, “We can’t use the word ‘agile’ around here any more because, for our stakeholders, that implies chaos where we only plan two weeks at a time.”
When people say “agile doesn’t work here,” it’s tempting to try to convince them they’re wrong. But there are three situations where they may have a point. We’ll share those three and how we’ve had a big breakthrough in how to address them.
“We’re in a domain where the cost of iteration is high”
In regulated industries, hardware manufacturing, and safety-critical systems, you can’t just “pivot” mid-stream. Changing direction after you’ve started FDA submissions, ordered manufacturing tooling, or begun safety certification is brutally expensive.
The agile mantra of “fail fast” becomes “fail expensively.”
Traditional project management promises to solve this with upfront planning. But that approach collapses when reality doesn’t match the plan—which it rarely does for anything new and complex.
Better to resolve uncertainty *before* committing to expensive execution. Do structured experimentation to answer critical questions while the cost of change is still low. Test key assumptions with prototypes, pilots, and proofs-of-concept before investing in full production.
In other words, be as agile as you can early in an initiative in order to resolve the core complexity and to get the data you need to able to plan more accurately. It’s not “no planning” or “plan everything.” It’s agility and planning where each serves you best.
“The business needs predictability that agile can’t provide”
Leadership wants detailed project plans with firm delivery dates. Other departments need coordination points. Customers expect commitments.
But we all know that most detailed project plans for complex work are basically fiction.
Instead, earn predictability in a systematic and focused way. Instead of bottom-up estimates based on sketchy early guesses, use reference class forecasting with actual data from similar projects (as recommended by megaprojects expert Bent Flyvberg). Run focused experiments on the biggest unknowns. By the time firm commitments are needed, enough uncertainty has been resolved to make forecasts that might actually hold up.
“We can’t just change the requirements”
We were recently talking with a client about a legacy core business system modernization. The new system needs to do everything the old system did, plus new capabilities. Multiple teams, complex integrations, business operations that can’t be disrupted.
“We can’t just pivot when we feel like it. We need to coordinate across teams, plan cutover windows, and ensure business continuity.”
But even when the scope feels locked down, there are still big unknowns about what those requirements should look like and how to implement the migration.
The opportunity is figuring that out while course changes are still possible. Before committing to specific technical approaches. Before finalizing the integration sequence. Before locking in timelines that assume everything will go smoothly.
Here’s what actually works
Distinguish between what’s actually complex (needs experiments) versus what’s complicated (needs analysis). Resolve the complexity through structured experimentation, then use analytical planning for the predictable parts.
This is CAPED: Complexity-Aware Planning, Estimation, & Delivery.
It’s not about throwing out planning discipline. It’s about earning good plans by resolving uncertainty before making big commitments.
Long-time agilists will see: Agile’s not dead. We’re using agile practices where they work best and structuring it all in a way that focuses on what matters to business leaders.
Where to go from here
If these impossible choices (predictability or adaptability, planning or agility) sound familiar, there’s a way off the pendulum with our CAPED approach.
If you’re staring at an initiative where “agile doesn’t work,” schedule a call. We’ll talk through what’s actually complex in it and what’s merely complicated. Our initial calls often feel more like a free coaching session than a sales call.
Want to build the capability in-house? See our CAPED training and certifications.
Last updated