BIOnDemand: It’s free…. Why aren’t you using it ?
Since the YouTube video below got released last week I have had many people ask me “what’s your secret” and “how did I spot that in the data” ?
As I am sure you are all aware I am certainly no clairvoyant and do not possess the skills that Dustin Hoffman’s character had in Rain Man.
What I do every time I see a new data set or a whack of information is to try and apply the “I” in Business Intelligence that is truly lacking these days. My thought process usually goes something like this when seeing the new information:
- How can we represent this data in a different way to change the way people look at this information ?
- What disparate/external data source can I bring into this report to enrich the value of the information and add another dimension to the decision making ?
In the case mentioned in the YouTube video above it was very much a 1. situation as the customer had been staring at the same report for well over a decade and it was clear they never fully grasped the effect of the data that was being presented to them. They do say that no presentation is complete without a pie chart and suddenly, when the data that was always buried in a long winded auditors reports was presented in our trusty pie chart did their outlook change on the data.
The beauty about BIOndemand, besides the fact that it is free, is that in a very short space of time you can be slicing and dicing your static data and getting new insights into your data. As you can judge by the tone of this post, I LOVE this product, and use it as my “secret weapon” when ever I am going to see a new customer. How do I do that? Simple!
Most customers we deal with are listed companies and have audited financials/reports posted all over the internet. Before the meeting I always extract some of that information and then, on my iPad, during an introductory meeting show the customer the “art of the possible” with their own data which really is a powerful message. It really can be as simple as taking their balance sheet and showing them visually where the bulk of their liabilities lie… the permutations are endless !
To me you can never underestimate the power of customers seeing the information that they can relate to – the world has moved past the stage of the “How many green T-shirts were sold in California ?” demos.
So what I decided to do, and I limited myself to an hour to prove a point, was to get some random data off the internet and put it in a BIOnDemand instance and see how/where I could add value to give you an idea of my thought process….
As I know a lot of us consultants spend our time bouncing between airports I managed to find this site which has all the data based on how and why flights are on time through out the US.
The graphing looks like this:
I then downloaded their raw data which came in this format:
If you look closely in the “airport_name” column there is a lot of information there, namely:
- Airport Name
Other enhancements I made to the file were as follows:
- Split the above 3 items out ( this way we can analyze which is the most reliable airport by State/Town)
- Added a column for Season ( Winter/Summ/Fall/Spring) as this may affect reliability in certain areas
- Added a total column aggregating all the flight information. from my world it really does not help that 100 flights are late. If that was in relation to 110 then you not going to fly out of there but on the other side of the coin if that was based on 100 000 flights you would be less concerned.
Below are a few screen shots that I have taken in BIOnDemand to get an idea of the new added functionality and the multitude of ways that you can now analyze your information:
Pic 1: The warmer it is the more people fly
Pic 2: Planning for TechEd in Vegas – SouthWest have the most flights into Vegas – but also the most delays !
Pic 3: The top 10 most reliable airlines flying into Vegas in October for two years (2010,2011)
Pic 4: Weather delays, which I “think” are peoples biggest fear of flying, cause less than 1 % of delays
Pretty cool stuff… well I believe so … I could go on forever – but you will be glad to know I won’t 😉
Yes there are some negatives about the free BIOnDemand service if you have to take a closer look, such as:
- It does not handle negative values very well
- Size limit (but as I said it’s free so stop being cheap)
- Dates are a bit of a nightmare to work with for sorting etc etc
There are 2 main points that I want to stress before people get carried away with BIOnDemand and go replacing their Enterprise Data warehouse (EDW) with BIOnDemand – there is a product that begins with a Q that already tries to achieve this:
- BIOnDemand is an amazing POC tool that can be used to get business buy in and give them insight/vision into what is possible with their information. When I present this to customers I always show a picture of an iceberg with this being the very tip of it ! Garbage in = Garbage out so the correct infrastructure/architeture and governance need to be in place to ensure quality information. ( As a “challenge” to the detail orientated folk I have purposely had some “finger trouble” in the Excel sheet which skews numbers a lot and could have cost someone their job – coffee with me at TechEd for the first person who spots it 😎 “Terms and Conditions Apply and Excludes Travel and Jamie Oswald”)
- Running under BIOnDemand is HANA. I am one of the front runners in making sure my customers buy HANA for the right reason and use case and not buy it due to market hype. In this instance HANA really performs amazingly behind the scenes and really gives the BIOnDemand the real time feeling for the end user. At the end of the day I want my end users to not care/know about where the information lies and I have to admit that SAP have it spot on here with HANA powering BIOnDemand. ** UPDATE ** Thanks to Ethan Jewett for pointing out that the free version on BIOnDemand does NOT run on HANA which really questions why SAP brand the the OnDemand site with “powered by HANA”. His full write up on this can be found here.
Please feel free to contact me if you would like to play around with this dataset on BIOnDemand as I will happily share it with you. If you would like access to the data directly then click here
Really keen on your thoughts on this technology and I feel if you ensure that you pitch this tool at the right level and carefully explain where it fits in then it will become an awesome part of your arsenal