A few days ago I mentioned I was writing a series of posts for CIMA Insight, which is the monthly web magazine of the Chartered Institute of Management Accountants – an audience that knows Excel quite well but probably has yet to discover PowerPivot.
Given that this series is starting from scratch, with an introductory post, it is mostly “old news” for readers here. But there are perhaps some high level explanatory points that are new. For instance, this diagram I sketched to explain the traditional tradeoffs between spreadsheets and formal BI: – a decision that results in impractically high costs sooner or later:
PowerPivot, of course, provides a curve that shares the low startup costs of spreadsheets AND the long-term maintainability and robustness of formal BI. Literally, PowerPivot is the end of that “damned if you do, damned if you don’t” situation.
To read the whole (brief) part one of the series, please click here.
As always, I’m interested in your comments.
Somewhere it would be nice to see the quantified cost/time savings of a PowerPivot model over an Excel or BI model. You chart comes close to this, and I realize that “real” comparisons might be proprietary, but it’s hard to sell “the world” on PowerPivot without hard numbers.
Got here after reading your post on CIMA Insight.
I agree with David on the need for hard data. The avalanche of false or at least overly optimistic claims made about process improvements and cost reductions by finance software providers in general makes that I am rather weary about any claims of long term cost reductions.
Maybe your next posts will help convince me:)
In the mean time, I will try to get a virtual copy of Excel running on my Mac to check out PowerPivot (since it doesn’t seem to run on my Mac…)
Rob,
I can’t resist to comment. It looks like you are suggesting PowerPivot as a replacement of organizational BI. This is like recommending Microsoft Access because SQL Server has higher costs. I think we should view and position personal BI (PowerPivot) and organizational BI as completing and not competing technologies. For example, PowerPivot doesn’t do much to address the proverbial issue with data quality and integration which is where 50-70% of the effort goes for implementing an organizational BI solution.
Also, are you suggesting that PowerPivot fulfills all BI needs :-)
Teo, of *course* you can’t resist comment! I’ve been going out of my way to offend traditional BI pros for nearly two years now! Frankly I’m hurt that I don’t catch more criticism than I do, thanks for dropping by :)
I have a long-pending blog post in the pipeline dealing with this very topic. But here’s a preview: I think you will agree with half of what I say in that post and possibly be revolted by the other half. Data warehousing and business intelligence should be considered two separate fields IMO, and even data warehousing methodology should be significantly changed. On net I think that means more work for you guys than ever, but much more heavily slanted toward SSIS/SQL and much less slanted towards SSAS/SSRS/etc. – true intelligence cannot be effectively outsourced to IT or consultants, and the only reason it has been outsourced historically is that there were no good tools for the business unit. The low satisfaction rate with BI projects is testament that something is broken. But not for long :)