In recent years I have had a relatively high number of project management engagements in the Data Warehousing and Business Intelligence domain. I was therefore quite excited to learn that Ralph Hughes, the author of Agile Data Warehousing is giving a talk in Melbourne.
Having managed (non-DW) Agile projects I was interested to learn about the opportunities to bring agility to the usually complex, long and expensive Data Warehouse projects. More specifically I was eager to understand how incremental deliveries can be deployed in a domain that requires substantial infrastructural investment before any tangible results are visible to the end-user.
Ralph is an eloquent presenter and his talk, in front a packed conference room, was well rehearsed and full of conviction. Ralph presents him self as somewhat of a convert. Having come from a ‘traditional’ project management and having been frustrated with the complexities and constraints imposed by the waterfall approach he has discovered Agile. This has re-kindled his professional enthusiasm and now he is a strong advocate and a practitioner of the Agile method.
Few comments about the content:
Agile advocates like to make references to evidence suggesting that non-Agile projects are failing at a fast pace, while Agile projects are better positioned to deliver successful outcome. It is not surprising, though utterly disappointing, that some of them, including Ralph in his presentation, rely on misleading information to strengthen their point. I was particularly put-off by a slide based on the notorious Standish Chaos Report. Ream of bits and bytes have already been devoted to demystifying and discrediting this report and the fact that Agile advocates need to rely on such unreliable source of information is beyond me.
I was excited to learn that large organizations with large Date Warehouse projects have successfully used Agile to deliver successful business solutions. What I missed, though, was a clear and unambiguous statement on the circumstances where Agile would not work. It is inconceivable to assume that Agile will work in all organizations, and in all circumstances. The entry and exit criteria for adopting Agile need to be better spelt out so organizations can better judge the suitability of this approach to their own local conditions.
In all fairness the 45 minutes talk was too short and was only suitable as an appetizer. Some of the time spent on highlighting the benefit of Agile vs. Traditional could have spent on looking at other, more technical aspects, in a bit more detail. I guess to get to that level of detail though you will be expected to enrol to one of the training sessions promoted as part of the presentation.
When time permits I will get myself a copy of Ralph’s book and no doubt a book review will follow.