Categories

A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

Some tweaks in the Scrum process so as to generate an overall schedule

The Scrum process can seem pretty radical for somebody who has not done this before. If you consider my case, I was used to normal waterfall or modified waterfall, where we would start work on a set of features, move onto more features as succeeding features got done, and so on. However, this resulted in a huge set of bugs as we kept on moving on, and soon we reached a stage where the battle was all about reducing the number of bugs in order to get a stable product. At some point, once we got the results of the usability analysis, it became clear that incorporating such feedback was always a challenge under the current system.
And then we moved towards using Scrum, given that other teams had adopted it, and reported higher overall levels of satisfaction, as well as made the bug curve a bit more straight preventing that huge bug load from catching on. There was also flexibility in the system to plan for incorporating comments from beta users as well as from external usability testing. Further, we ended up giving more time to product management and the user interface teams to give requirements and plan the workflow, making the end product much better.
However, the biggest challenge faced was towards the initial assessment about the number of features we could implement. To make a mark for the product, delivering some new features is critical. In the past, we would do some initial heuristic + some development analysis estimates to generate the list of features that can be done in the entire schedule, since that list was needed for getting approval for the release of the product. However, in Scrum, one does not have this initial estimation against the features, and hence we did have a challenge of being able to not define the exact number of features that could do.
So what did we do ? A compromise, which was not 100% accurate (more like 60% accurate); we did do an initial estimation and scheduling of the more important features (and this did take around a week to complete), enough to generate the feature list that we wanted. While none of us was really comfortable doing this initial analysis rather than depending on the Sprints to work through the Product Backlog, it was something that we needed to do. Given the need to do an initial estimate of the probable features, not sure whether there are any more solutions.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>