By: Chris Sims
Today finds me in Seattle for the Agile Open Northwest conference. This morning I lead a session called "An Experiential Introduction to Agile" which was based on our Experiencing Agile mini workshop. I invited participants to share their insights on sticky-notes, with a promise that I would post them. So, here they are.
Cheers,
- Estimation works better when done in smaller chunks.
- Easier to fix issues if we discover them sooner, even in mid-iteration.
- Success begets success
- We have been estimating at the beginning of our projects, when we know the least that we will ever know!
- Working in an agile, single-piece-flow, we saw progress as we went. The felt better and we could work slower and yet accomplish more.
- Feedback is more useful when it arrives early, instead of at the very end.
- Going through a full cycle, quickly, facilitates more learning
- When we worked 'waterfall style' we had lot's of work-in-progress which became waste.
- Taking things (stories, features, units of work) all the way to completion helps us estimate future delivery. Completing a 'phase' as in waterfall isn't nearly as good for this.
- It's really hard to know how much we will be able to get done in advance, even for simple tasks. Much better to do some and then see.
- Requirements –> Completion Since each phase in a waterfall is different kinds of work done by different people, the rate of progress for one type of work doesn't have any bearing on the timelines of other steps/phases.
- The exercise was a wonderful demonstration of how easy it is for even experts to fail. No matter how good the 'big plan' it's probably wrong. Better to work incrementally.
- I was surprised that even a basic experiential simulation like this can bring out all the real emotions of a real project.
- Basing our schedule on the original upfront estimate lead to much more waste, than adjusting our plan as we worked.
- People gain satisfaction from shipping
- Get customer feedback early and often!
- Not having to do try to deliver on unrealistic estimates makes for a more relaxed and productive environment.
- "You know the least you ever will about your project when you begin."
- Taking small units from beginning to completion was much more efficient that working in big batches. We also caught bugs sooner, which is less costly to the project.
- Incremental tasks –> opportunities to learn how to do it better.
- Delivery in small batches allows customers to see product early, if they have changes those can be applied with less waste.
- Moving something from concept to completed & shipped more quickly, caused me to get more done and also gives me more satisfaction.
- Shipping in smaller increments felt better, a greater sense of accomplishment.
- Making a mistake on 1 item is easier to fix than on 20 items. Much better to get the feedback early!
- We make all of our estimates at the beginning of the project, when we know the least ever. Worst time to estimate!
- We shipped much more overall when we delivered in small increments.
- Customers don't always know what they want until they see something.
- Somebody is spending real money when you don't deliver.
- My team needs to figure out how to test early with no QA resources available.
- Psychological effect of incremental shippable production helps sustainability.
- Shipping all at once rather than small increments caused a lot of wasted work that was never delivered.
- Estimate by seeing how much we can do, as a complete task, in a time-frame.
- Slow down, think about it!
- Team morale was much higher when we took the 'agile' approach.