Faced with the current tough economic conditions, can we predict how organizations might react?
(1) People will try to do without BI — and fail
BI is important, but rarely urgent. In dire economic conditions, some organizations will be too busy trying to survive to think about doing analysis. But action without analysis is called guessing, and is unlikely to help.
(2) People will revert to hand-coding and excel macros — and waste a lot of money
Corporate cutbacks, "thou shalt not buy anything" policies, and new levels of sign-off will encourage some people to attempt to do analysis without extra software investment: hand-coded data extraction in SQL, data manipulation using Excel macros, etc.
Over time, the work involved in developing and maintaining these solutions will cost much more than purchased packages. The scripts won’t include any comments. The people responsible will leave the company exactly 4.3 months later. Nobody else will be able to figure out how it works. The project will limp forward for another 7.7 months, and then vendors will be invited in to do prototypes which will reveal that the scripts are faulty and the company has been basing their decisions on incorrect data for the last year. This will be covered up.
(3) Organizations will implement standards — but omit to change organizational structures
Keen to reduce costs, organizations will standardize their BI environments — but will balk at the perceived cost of implementing a dedicated central BI organization. The result will be lower procurement costs, but without a BI competency center there will still be silo BI projects. This will result in continued needless duplication, BI skill shortages, multiple definitions of¨"profit", "headcount", etc. Savings will be a fraction of what they might have been, with no improvement in overall view of information across the organization.
(4) Business units will find it easier than ever to implement their own solutions — to the detriment of the company as a whole
Chafing against corporate BI standards that they didn’t chose, business units will find it easier to ever to implement their own "shadow" BI systems. Lacking any incentive to plan how their system fits in with the others, the result will be more silo BI systems, making it harder than ever to get a "single version of the truth" across the organization.
The answer? Step one: BI organization
What can organizations do about this? Now more than ever is the time to implement BI shared services or a BI competency center. Create one by bringing together all the resources that are currently being wasted collecting and analyzing information separately in each department. The first goal of the team should be to prioritize BI projects and consolidate existing projects and solutions, eliminating waste and increasing information flow. Managed correctly, the group will more than pay for itself.
Comments
5 responses to “What Might Go Wrong in Business Intelligence in 2009?”
Peter, thanks for the comment — I completely agree… I often do presentations on setting up a BICC, and the hour time slot is never enough time to even scratch the surface on what can go wrong and how to prevent it…
Your conclusion that BICCs are the way forward is a good one. However the challenge is not one of logic, but one of structures. A BICC by definition cuts across old lines of control and leads to organisational change. Pet projects have to be dropped, staff have to be realigned, software needs to be rationalised. All of this can be difficult to achieve, particularly across different countries and business groups and particularly where central control is weak. Finding a way to address these issues is the best and only way forward.
Although I totally agree with you in terms of corporate skepticism in investments, here in the Middle East, specially the rich oil states, business is booming, we are continue to slow down company requests to deliver everything all of a sudden. There is a huge appetite and hunger for BI in this region, thanks to the unshaken oil economies
I totally agree with your point #2 – especially crucial are data integration solutions in automating SQL ETL/ELT. I also believe that the overall the Business Intelligence market will survive the downturn. For more of what I said and the track-back: http://blogs.oracle.com/dataintegration/2008/12/bull_market_on_bi_in_2009.html
Bull Market on BI in 2009
Recenlty Timo Elliot, wrote four top reasons for “What Might Go Wrong in Business Intelligence in 2009”. In his reasonn #2 he cites: Corporate cutbacks, “thou shalt not buy anything” policies, and new levels of sign-off will encourage some people…