Should I use SCRUM for projects lasting 4-8 weeks ? (building websites)
Hello - previously I was using a SCRUM in the company where we were developing our own product within few Scrum teams - I would say the life was easy.
Now I am working with a small projects like 3 (2 weeks) sprints.
I am just wondering should we use SCRUM at all in such a projects?
Just to add that team size is not an issue here.
Doubts I have here:
1. Once the contract is signed there is no much time to plan/prepare everything in front of the sprint start - there is even no time to do a Sprint 0 - because the budget is small in such a case.
2. Not much possibilities to inspect and adapt - especially after 3rd sprint in mentioned example.
3. Very frequently it happens that project starts and e.g. designs are not ready - so in fact the requirements are not ready - what to do then ? Should we base our User Stories on some assumptions ? or simply wait till we will receive approved designs ?
I will be very grateful for any support here :)
Tomek
You have valid concerns. Here is what I would say about them.
- I don't really agree with Sprint 0 concept. Scrum says that you should deliver incremental value with every sprint. Sprint 0 doesn't really deliver any incremental value. The discovery/knowledge you mention should come via refinement and discussions among the scrum team and stakeholders. I would say that it seems very difficult for you to actually deliver any value given that you don't really understand much about the work. But then it also sounds like you have a struggle on your hands to complete the work no matter what "method" you end up choosing.
- You have as much opportunity to inspect and adapt as you would in any Scrum project as long as you don't remove any of the Events or neglect refinement. And just because your contract is over does not mean that your team shouldn't do a retrospective on the sprint. What is learned may not apply to the next sprint but it could apply to a future project.
- Build your stories based on what you know right now as empiricism demands. If you learn something new the next day, inspect and adapt. Even if you had all the designs done up front, you will learn things that could influence changes. There is nothing in Scrum that says designs have to be completed upfront. In fact, most Scrum practitioners agree that the designs should evolve naturally as you build, review, inspect and adapt based on stakeholder feedback and experienced gained.
Now let me add a couple of questions.
- Does your team already know Scrum or will they have to also learn Scrum in 6 weeks? Remember Scrum takes time to learn and understand. It could be overkill for that short of a time unless you also plan to continue using it.
- What benefit do you feel would be obtained by using Scrum? Is it just because you liked using it previously? While Scrum is a good tool for software development, it isn't always the best one.
Honestly, unless the people working on the project are already a functional Scrum team, I'd consider something else.
I am just wondering should we use SCRUM at all in such a projects?
Is there much complexity, risk, or uncertainty to be brought under control when building these websites? If you are confident that an acceptable product can be developed in 4-6 weeks it suggests there might not be. If there is, would having a regular Sprint Goal to aim for help?
Hey
@Daniel - regarding the Sprint 0 I agree - all these could be done up-front on the refinements.
Answering questions:
1. Yes - people know already a Scrum way of software development.
2. Thirst thing is that people know the Scrum way.
So the question here I have - what else should I use for such projects ? Kanban ? Anything else ? Any idea ?
@Ian
There is a risk and uncertainty - but unfortunately within 3 sprints plan something which has changed(on customer feedback) during Sprint 2 to Sprint 3 is quite late - for me I would need something where I can really quickly adopt... Maybe better would be shorter sprints ? or simple Kanban... I do not know...
For sure I will put more focus on the sprints goals - that was a good hint in fact here - thanks @Ian :)
@Ian has a very good point. The Sprint Goals will help the team focus on delivering.
Let me remind you that Scrum does not limit you to delivering only at the end of the sprint. If you have a working increment that is potentially shippable and provides additional value 3 days into the the sprint, you can get feedback from the end user at any point. In fact, you should be getting feedback constantly. And the Development Team can introduce changes to the sprint backlog as long as it is to adjust based on feedback or things learned and it does not endanger the Sprint Goal. They are not committing to only doing the stories they start with. They are committing to a forecast that will deliver the Sprint Goal based on what you know at the time of Sprint Planning. If things are learned during the execution of the sprint, work as a Scrum Team and adjust accordingly. Just try to do it controlled and in a manner that will not prevent you from delivering.