How to be successful at BDD Specifications Workshops?
- by sigo
Today we tried to introduce BDD in our software development process by having a specification workshop.
For this workshop we had 2 developers, 1 tester and 1 business analyst. The workshop lasted 1h30 and by the end of it we managed to figure out some BDD scenarios for our new feature.
We tried to focus on finding the scenarios that we could miss, and the difficult ones.
At the end of the workshop some people were actually unhappy with the workshop.
One developer felt he wasted his time as he was used to be given out the scenarios directly by the business analyst and review them with her.
The business analyst didn't feel confident with our scenario coverage (Had a feeling that we could have missed out other important stuff) but more importantly felt that this workshop was also a waste of time as she could have figured out all these scenarios by herself and in a shorter period of time.
So my question is how that kind of workshop can actually work.
In the theory, given you have a new feature to develop, you put the tree 'amigos' (dev/tester/ba) in the same room so that they can collaborate together on writing the differents requirements for the new feature using examples. I can see all the benefits from that. Specially in term of knowledge sharing and common product/end goal/done vision.
But in practice, we still think it is more cost effective to first have a BA to work on his own on the examples and only then to have the scenarios to be reviewed/reworked by the 3 'amigos'. By having the BA to work on his own, we actually feel more confident that we are less going to miss out stuff + we still get to review the scenarios afterward to double check. We don't think than simple brainstorming/deliberate discovery is actually enought to seriously cover all the requirement for a feature. The business analyst is actually the best person for that kind of stuff. The thing we just do is to review what she wrote and see if then we have a common understanding (which could then lead to rewrite some of her scenarios or add new ones she could have missed).
This workshop lasted 1h30, and by the end of it, we didn't feel confident enought about wha we did...sure we could have spent more time on it but honestly most people get exhausted after 1h30 of brainstorming.
So how can you get that to work effectively in practice ?
