This post is actually an edited version of something I posted a few days ago on LinkedIn, but I liked it so much I decided to post it here as well. Basically it explains the philosophy behind the creation of Vizubi.
1) The intelligent part of BI is people. It doesn’t matter what kind of software you have, unless you have people who can turn data into actionable information BI is pretty much useless because it won’t lead to good decisions.
2) The people who have those analytic and decision making capabilities rarely have advanced technical skills and it probably isn’t a good idea to require them to learn those skills to do their jobs (their jobs being turning data into actionable information).
3) Excel is pervasive, “within reach” of the technical skills of most people in any organization, relatively cheap in terms of both licenses and training necessary and well liked by most users (if not by IT). Deployment time is usually low because Excel is already available; it also has the functionalities which allow analysts to do lots of analyzing. Most importantly, whether we like it or not, Excel (or an excel worksheet embedded in a word document) is currently the de-facto last step in most reporting processes because it is used by junior analysts and C-level managers alike and this is likely to remain unchanged in the near future.
4) Currently deployed desktop and laptop computers have lots of computational power. Using new data management techniques (column based, in memory and associative are all words that come to mind), these machines can handle huge amounts of data in useful ways (sorting and filtering millions of records in seconds).
5) To do good BI you need good, clean data otherwise you just get “garbage in garbage out”. IT has to make sure that the systems (typically ERP plus other stuff) capture, store and make available clean data. In addition, it is important that users have access to all the data at a low level. This will allow them to both contextualize the data through filtering and discover relationships that might not have jumped out when using aggregated data.
If organizations are able to get IT to give users direct access to data and there are easy to use tools to do that through Excel, then users will cut out the layers of BI that stands between them and data and do analyses directly through Excel plus some tool. It’s faster (low deployment time and easy prototyping), easier (analysts know the tools), relatively inexpensive (Excel is already available and training is not required), and most importantly, it gives analysts direct access to atomic level data which should lead to more interesting results.
That is the reasoning that has led us to develop Vizubi.