Business intelligence software has been with us for some time and we’ve seen many new technologies enter the BI stage. There’s no disputing that BI can offer tremendous benefits as an IT investment, but the market is changing fast and providers (note the BI Giants) need to gear up for the next generation – which goes far beyond the latest macro trend.
Driven by the need to cut costs and make room in the BI budget for new investments, Generation-Why, the ‘query generation’ will ask tough questions and have an insatiable need not just for data or insight, but for fast answers.
More than ever, the focus will be on response times: a recent survey from BARC revealed that 38% of users are unhappy with the performance of their SAP BI modules.Thanks to “In-Memory Computing”, SAP may be aiming to increase BI response times by up to 100%, but others, particularly the smaller niche players, can still win the race – both with a speedier and more affordable option.
I see the future for speedy business application answers not in the central processing unit (CPU) as SAP does, but in the graphics processing unit (GPU). Modern graphic cards have 100s of special processors that can be used for parallel computing: an approach that the Universities of Freiburg and Western Australia are helping us research. Intel and AMD also need to think anew.
Alongside In-Memory-Analysis, many new technologies have entered the BI arena. Many call this BI 2.0, but what does this really mean? At the end of the day to most of the new BI vendors it means more fancy charts, graphs and dashboards – with some larger fonts, brighter colours and rounded corners thrown in for “interactivity”. All this is well and good, but this is just pure marketing if it doesn’t help the user make better decisions. If this BI approach is to be seen as the “next generation” then it needs to go further.
BI applications and tools need to be rooted in the workflow – and should be cognisant of the types of decisions that need supporting. BI needs to be aware of the domain context – i.e. which industry, which department. Because without this, the best BI can do is to provide pretty visuals and hope and pray that the user knows how to translate them into intelligence. This is not the Jedox way, as usability and business value need to take first priority…..new technologies alone with little value add will not make customers satisfied.
Very often, it’s the smaller BI operations and niche players who are often the innovators here. Most of the BI-g Boys haven’t kept up with the market.
Looking at the BI tools and software available, one has to ask the question, what’s the next generation really looking for and are we ready? And we don’t need expensive answers here, but super fast ones. In an ideal world, Open Source answers.
Mr Kristian Raue,
You speak my language and it aint’ German.
Regards
Peter Fischer
Quote
BI applications and tools need to be rooted in the workflow – and should be cognisant of the types of decisions that need supporting. BI needs to be aware of the domain context – i.e. which industry, which department. Because without this, the best BI can do is to provide pretty visuals and hope and pray that the user knows how to translate them into intelligence.
Indeed, We have done some data mining projects that are part of decision nodes inside workflow and it works and helps the employees.
On another note, I totally agree with Mr Kristian Raue, “it doesn’t really matter the type of BI suite to be implemented “, if the Consultant(s), Employee(s) are not smart enough to translate into cubes, charts, reports into a sustainable solution to help increase the competitiveness of the company everything will be useless.
Quote
BI applications and tools need to be rooted in the workflow