News for hackers. Stuff that matters.

A comment on Skateboard, Bike, Car: Building Products the MetaLab Way.

Re:Skateboard, Bike, Car: Building Products the MetaLab Way by aethr on Tuesday October 25, 2016 @10:06PM (#12793364)
As someone at in an agency still working in a "waterfall style that was popular in the 90s", how do we convince clients with a fixed budget and a list of "must haves" to go agile? We have PMs with agile experience, all the developers are onboard, but when every other agency is promising to deliver what the client needs for a fixed cost, why would they take the risk?

We've had some small successes starting agile projects with existing clients who trust us, but articles like this make it seem like we're incompetent for even engaging in a waterfall project in 2016. When you're pitching for new work with a client who doesn't know the difference between waterfall and agile, one seems like a promise and the other seems like a risk. As software engineers, we know the opposite is true, but why should they trust us?

[ Parent | on HN ]
Re:Skateboard, Bike, Car: Building Products the MetaLab Way by repeek on Wednesday October 26, 2016 @09:30AM (#12796270)
When you say "go agile", what do you mean? Why does the client care how the project is delivered?

If the client wants to pay a fixed bid then bake that added risk into your estimate.

For clients - if they're asking for something new/unique, what they have in their minds at the beginning is just a theory and needs to be tested -- working in iterations and focusing on "how do we help you prove your hypothesis in the most time/cost effective manner". Agile is about incremental, demonstrable progress to provide clarity and reduce risk.

[ Parent | on HN ]
Re:Skateboard, Bike, Car: Building Products the MetaLab Way by popdoit on Tuesday October 25, 2016 @10:59PM (#12793564)
God, I feel the same way. I've spent most of my career at agencies and, yes, many of them are still using waterfall project planning.
[ Parent | on HN ]