Search "is waterfall project management dead" and you'll find a discussion thread from 2017 asking the exact same question. Search it again and you'll find an article from 2022 declaring waterfall dead with an exclamation point. Search it again and you'll find a Substack post from December 2024 insisting waterfall isn't dead, actually, and still thrives in modern project management. Search it one more time and you'll find someone in 2025 revisiting waterfall's "enduring relevance" as if this were breaking news.
This is not a trend. This is a genre. And it's been running, largely unchanged, for close to a decade.
So instead of adding entry number eleven to that pile, here's the actual answer - with the data to back it up - so the next person searching this question has somewhere to land that isn't another rehash.
Why This Question Keeps Resurfacing
The honest answer isn't complicated.
Waterfall never went anywhere, agile advocates have spent a decade predicting its demise, and every few months someone rediscovers that predictions and reality have diverged, and writes it up as news.
One recent piece even frames it as a genuinely controversial take - asking readers to consider whether they've ever actually followed a rigid, by-the-book waterfall process, and using the answer as evidence that waterfall "no longer has a place in our ever-evolving world.”
That argument has been made, in some form, in nearly every "waterfall is dead" post for the last ten years. It's compelling rhetorically. It hasn't been true empirically at any point in that decade.
Meanwhile, the "actually, waterfall still works" counter-genre keeps needing to be written because the first genre keeps needing to be rebutted. Construction, aerospace, and healthcare specifically get named over and over as the industries where waterfall remains indispensable - not because anyone's being nostalgic, but because those industries have compliance and documentation requirements that a linear, phase-gated process serves well.
Ten years of the same argument being made and rebutted is not a live debate. It's a stalemate that content calendars keep mistaking for news.
What the Actual Data Says
Here's where most entries in this genre stop short - they argue from anecdotal or personal experience rather than looking at usage data.
So let's look at it.
Standish Group's CHAOS database, which has tracked over 50,000 projects across three decades, shows overall project success rates remain below 40%: 35% of projects succeed outright, 46% are "challenged" (completed but over budget, over time, or with reduced features), and 19% fail outright.
Within that data, agile projects do show a meaningfully higher success rate (42%) than waterfall projects (26%) - a real data point, and one the "waterfall is dead" camp is right to cite.
But that is success rate on a like-for-like basis, not usage rate. And on usage, waterfall hasn't disappeared - it's still one of the most commonly chosen approaches, not a relic.
Reference: Standish Group Chaos Report via: https://agiled.app/statistics/project-management-statistics
PMI research has found that roughly 44% of projects use the waterfall method while around 30% use agile methods, with the remainder using a hybrid of both. That's not a methodology in decline. That's the single most common approach still in active use, industry-wide.
Reference: PMI Pulse of the Profession: https://www.pmi.org/-/media/pmi/documents/public/pdf/learning/thought-leadership/pmi-pulse-of-the-profession-2024-report.pdf?rev=c480c0b72ee8466eaba10132b614c5d7
The two data points aren't actually in conflict - they're answering different questions. Agile projects succeed at a higher rate when both are measured the same way. But that doesn't mean waterfall projects are rare, poorly chosen, or on their way out. It means agile tends to outperform waterfall on average, in the aggregate, across all the contexts both get used in; which is a different claim than "waterfall no longer belongs in anyone's toolkit."
Why the Aggregate Comparison is Misleading
This is the part nearly every entry in the genre skips: averaging success rates across industries erases the reason waterfall gets chosen in the first place.
Nobody selecting waterfall for a bridge, a hospital wing, or a regulatory-compliance system is choosing it because they haven't heard of agile.
They're choosing it because the project has fixed, well-understood requirements, heavy compliance and audit obligations, and low tolerance for the kind of mid-project pivots agile is built to accommodate.
Comparing waterfall's aggregate success rate to agile's aggregate success rate, without separating by project type, is comparing waterfall's performance on hard, compliance-heavy, low-flexibility projects against agile's performance on projects that were suited to iteration in the first place.
That's not a fair fight, and it's not what either data point is actually measuring.
This is the same point PMI's own performance data makes about methodology broadly: once you account for what the project actually needs, performance gaps between approaches shrink dramatically.
The CHAOS success-rate gap isn't evidence waterfall is failing - it's evidence that waterfall gets deployed on harder, less forgiving projects on average, and still lands the plane 81% of the time (35% success + 46% "challenged but delivered").
So is Waterfall Really Dead?
No. It's one of the two or three most commonly used project management approaches on earth, it remains the default in industries where documentation, compliance, and sequential dependencies aren't optional, and roughly 3 in 10 projects globally still run on it as a primary methodology.
It is not niche, not fringe, and not fading in any way the data supports.
Is it the right choice for every project? Also no - nobody serious claims that, including PMI, whose own research is what most "waterfall is thriving" posts lean on to make their case.
The honest, boring, non-clickbait answer is the one nobody wants to write about: waterfall is neither dead nor universally superior.
It's a tool that's well-matched to a specific, still-common category of project - and the fact that this needs restating every eighteen months says more about the discourse than about the methodology.
The Real Question Worth Asking
If you're choosing a methodology for your next project, "is waterfall dead" is the wrong question entirely.
The right question is the same one we've argued elsewhere on this blog: Does your project have fixed, well-understood requirements and low tolerance for iteration, or does it need room to adapt as you learn?
Answer that, and the "agile vs. waterfall" debate mostly answers itself, without needing a decade-old argument to settle it for you.
Buena Suerte,
