Agile vs Waterfall: Waterfall Wins! + CHEAT SHEET

It's easy to fall into the trap of thinking that Agile is better that Waterfall. What needs to be be true for Waterfall to win over Agile?

= = = = = = = = = = = =

New for 2024: my best-ever training:
"How Your Agile Teams Can Achieve Predictability and Productivity WITHOUT Burnout"

= = = = = = = = = = = =



- My process was Waterfall
- My dad's process was Agile

My dad's Agile process won by a landslide. But why did it win?

And are there cases where Waterfall would be victorious?

-------------------
40. Agile vs Waterfall: Waterfall Wins! + CHEAT SHEET
# #DevelopmentThatPays

Previously... My 19 year old self took on my dad at the classic DIY challenge of hanging a shelf. To no one's surprise, my dad won. It was not just a victory of experience over enthusiasm. It was also a victory of Agile over Waterfall. Was Waterfall destined to lose Or could it have claimed victory What has to be true for Waterfall to win Shelving ----- If you missed the last episode... ... you may decide that you were glad that you missed it. Because we spent the entire time attaching this shelf... ... to this wall. Not once... but twice. My method of hanging the shelf - very much a Waterfall approach - scored high on efficiency: I measured everything marked everything drilled everything screwed in all the screws But it scored badly in effectiveness: the result was a wonky shelf. My dad's process was was much more iterative - much more Agile. He marked one hole He drilled ONE hole He fitted one plug He fitted one screw [And so on...] He checked for level multiple times, so it came as no surprise that the end result was also level. It was a fair fight: my dad's method won fair and square. But it did make wonder: why did it win And could there be situations where a Waterfall approach would be victorious Let's change the parameters a bit: instead of the challenge being to hang a shelf, the challenge now is to hang 20 identical shelves. My dad's process is awesome for a single shelf, but for 20 shelves it's far too much work. His process is bespoke. I need something more "cookie cutter". I've come up with this jig; I think it's going too make all the difference. Match this mark to the centre mark on the wall Adjust for level... using the built in spirit level. Press it firmly against the wall: the rubber backing keeps it in place. The drill guides ensure that the holes are in exactly the right location. It also ensures that the hole is drilled at right angles to the wall. Drill all 4 holes. Remove the jig 4 plugs in Offer up the shelf In go the screws - but not too tight. Final check for level. Tighten the screws all the way. Not just a job done well. But a job done quickly. Waterfall just beat Agile. Why did Waterfall win Clearly, the jig was the enabler here. But in order for the jig to work as expected, at least two things must be true: All of the walls must be reasonably flat if they're not, then the jig won't "stick" to the wall. All of the shelves are identical if they're not, the holes drilled in the wall won't line up with the holes on the shelf. Notice that neither of these conditions is necessary for my dad's process to work: his process will always work: on flat walls; on bumpy walls; with shelves of all shapes and sizes. So it would appear that it's all about predictability. If the steps can be predicted with a high level of confidence for example, if it's an activity that we've completed many times before - then a Waterfall process is likely to be the best choice. Conversely, if the level of confidence is low for example, if we're attempting something for the first time - then an Agile process is going to pay dividends. Agile and Waterfall. Horses for courses. Many thanks for watching. Talk to you next time.

[anindividual]

This makes a lot of sense. While one might use an agile process for their first product, when you start to make similar products which have the same basic ins and outs, you can waterfall it so fast.

[tigerli]

The level of risk determines the better way. Good summary.

[MikeJonesTechno]

Nice explanation, but I feel this is also missing something important? Context of batch v single piece flow, and the concept of continuous improvement. OK, so that's two things.

"Put up 20 shelves" - compare agile and waterfall...

Agile will have learnt through continuous inspection / retrospectives to improve process and create the jig - perhaps after the second shelf 'there has to be a better way'.

Waterfall would have planned 20 shelves as a single waterfall project - plan everything, measure everything, drill everything, fit i.e. carefully measure and mark up 20 walls, then drill holes for all 20 shelves - the same traditional way as your younger self - and eventually offer up 20 shelves to discover all 20 shelves are wonky. Then a post implementation review would have blamed each other, the tools, or the walls for the mistakes. Perhaps somehow the waterfall would figure out to try a jig.

Waterfall would then try the jig on the next 20 shelves. Drill 20 shelves, fit 20 shelves. Only to discover the jig isn't 100% square, or the drill isn't always going in 100% level. Holes are in the right spot but each hole not always at the right angle.

In your example, how did you design and create the perfect jig? 20 shelves at a time? one shelf at a time? one screw at a time?

Through agile of course! Agile would continuously refine and improve the jig after every drilled hole.

1. Make a jig. Try out the jig on the next shelf, but ONLY on the top 2 screws. This is the agile bit!
Did the jig help get the top two screws correct and level? If not, get hammer to adjust and correct the shelf as before.
2. Make small adjustment to the jig. Try the modified jig on the next shelf, but ONLY on the top 2 screws. This is iteration 2!
Did the jig help get the top two screws correct and level? If not, try another small adjustment to the jig and try again!
3. Continue making small adjustments - plan - do - check - act - to create the correct jog for 2 screws, then 3, then 4 screws.

Perhaps initially I try a paper or card jig to mark out correct screw holes eliminating my incorrect measurements and ruler skills. Perhaps a paper or card jig isn't robust enough and flops around or rips easily. Maybe I try a thin chipboard or MDF template.
Maybe the chipboard is too thin, holes are correct place but I'm still drilling at the wrong angles (lack of experience with a drill).
Maybe I add 20mm blocks to the jig to help guide the drill nice and straight. Perhaps I permanently attach a mini spirit level to the jig to take one step and one tool out of the process. Perhaps I add rubberized backing to make it easier to hold the jog firmly in place against the wall.

I try each improvement one at a time by learning and adapting.

Agile would create the perfect jig in perhaps 8 shelves.
Waterfall would create the perfect jig in perhaps 8 projects of 20 shelves - 160 shelves.

Agile wins in the jig making contest?

Waterfall does win in the cookie cutter identical process - that you never want to improve - or at best only want to improve very slowly over several huge batches, and where quality is less important than delivery - without constant inspection you may not realize there is a defect in the jig, or a manufacturing defect in one of the shelves that wasn't exactly matching the template. You wont find out until the next batch of 20 shelves is done.

Agile wins at any time you want continuous improvement, or when quality is more important than delivery.

[Haasgen]

Great job!
Simple, quick, well explain and very nice animations!
Thanks!

[HassaanRaza]

love your way of doing things ! nice job.

[theguardian]

This is a nice example but there are many other reasons to user traditional project design, planing and exectution (aka waterfall), for instance instead of mounting a shelf try to build a bridge (or a car or something more complex) with agile and most likely the bridge will fall before it's even half way done.

[SchadenfreundeParody]

I'm no fan of Agile, but you ignored one huge moment:" making a jig" - where did you get the jig from? Making it first would be more time consuming and difficult (or even impossible), and if the task is changed? a new jig?

[shama-_-]

thanks a lot that help me can you send the link of this awesome presentation slide the template

[hskoppek]

Which method did you use to build the jig with? Agile or waterfall?

[adeemmmm]

The only thing is that u made the jig with probably an agile process..
So Agile wins 👑

[wewantthefunk]

The process of hanging 20 shelves using the jig isn't waterfall, it's manufacturing. After the first shelf is hung and the jig is created, you aren't requiring critical thinking. Hanging 20 shelves is a high mechanical, low thinking task. Figuring out the first shelf and the jig is high thinking. This analogy doesn't work, and waterfall never wins.