Oz Analytics

The Business Of Interpreting Data

Last night I created a short video for a talk I am due to give tomorrow at the Mastering BI With SAP conference in Sydney. The subject of the talk is "Why Bother With BI?" In the video I'm trying to make the point that analytics in the corporate world generally has two components: data and commentary.

Data can often be used to support both sides of an argument and the words we use can (and are) open to wider manipulation and interpretation than our common-sense leads us to believe. The video is trying to make this point in an interesting way.

As I said, I just made this and before I present it to a couple of hundred people, I thought I would publish it here and get some feedback. So please let me know if you like or hate this. You could be saving me some public embarrassment!

I should also give credit to the Argentinian politician whose campaign ad was my inspiration. He didn't get elected so he is to blame if my video doesn't work ;-)

DW 2.0 Asia Pacific Summit

I am happy to announce that the Business Intelligence Group is supporting the DW 2.0 Asia Pacific Summit to be held in Sydney at the end of October.

We are not only supporting the event but I am going to talk about 'Decision Mashups - Strategies for Combining Structured and Unstructured Data for Better Decision Making'.

Once again I make the offer for any readers of this blog, or members of the CORTEX to make contact so that we can meet up at the event.

Here is the brochure for the summit:

Download DW 2.0 Asia Pacific Summit Brochure

I also see that Eugene Dubossarsky from Presciient and moderator of our Forecasting Special Interest Group on the CORTEX is to speak on 'The Future Role of Analytics: Arms Races, Success Stories and the Changing Role of the Analys'. Eugene always has an interesting perspective and I'm looking forward to his talk.

Social Networking and BI

I was sent a link to a presentation on social media and BI by US/Indian company erada. They specialize in developing solutions that work in a SAP environment.

In the presentation they quote Gartner and their recent (February 2009) survey of 1,400 CEO's who reported that they did not achieve BVA (Business Value Attainment) in their BI investments  What's really interesting is that less than 9 % reached their desired BVA when they went live. BVA is broadly a measure for stakeholders of whether-or-not a good investment has been made.

What is not clear from the Gartner analysis that I have seen is why you would expect to see instant return on your investment. If you buy a truck for your company you expect it to earn a solid return for you over the future years - not instantly the day you buy it. Isn't this the same for BI (and any other software) investment. Or am I missing something?

I suspect Gartner is trying to measure the perception that CEOs have of what they are getting at launch versus what they expected to get (based on promises) when the BI investment decision was made.

This is an important point and I myself go to great lengths not to over promise to stakeholders. It's also not necessary to do so as the benefits of the BI investments (at least where I have worked) are simply massive. The trick is to use language your stakeholders understand. Delivering value also helps ...

Here is a copy of the presentation. Persist with the way the slides are built as there is some good information to learn:

Social Network BI-Outside In_sept09

I like the slide pictured above as they make a good point that just churning out reports each day does not deliver insight.

I also like the sound of their SmartCube product that (according to their website blurb) "helps in redesigning the SAP InfoCube to reduce query time, load time and footprint on BIA." Anything that helps InfoCube work faster gets my vote!.

Why Organisations Make Bad Decisions

I recommend reading the Psyblog entry discussing the value of dissent. It presents a well thought out argument as to why corporations make bad decisions. There are lessons here for those interested in better understanding the group context within which decisions are made in most large organisations.

Although it is not exhaustive, there are some tips on how to create constructive dissent. This is very difficult to do in practise - and doubly difficult if you are an analyst with the data already to hand that proves you are correct!

Here are Psyblog's ways to achieve 'good' dissent:

• Hire someone who genuinely disagrees with the group
• Hire leaders that facilitate discussion. 

Doesn't sound like rocket science, so why is it so difficult? Here's why:

• Organisations often recruit on the basis of who will 'fit in' and not 'rock the boat'. The stereotypical yes-man often emerges, perhaps unconsciously, as perfect for the job.
• Group cohesiveness is highly valued for productivity ('are you a team-player?'): groups who are always bickering are perceived as getting less work done.
• Disagreement and the expression of conflicting opinions makes people uncomfortable and they try to suppress it, partly because:
• Dissent is easily misinterpreted as disrespect or even a personal attack.
• Dissenters are often labelled as trouble-makers and targeted for either conversion to the consensus or outright expulsion from the group.

Thomas Davenport, author of Competing on Analytics: The New Science of Winning (Harvard Business School Press) makes a similar but much broader point. He advocates a new emphasis on people as the key ingredient in successful predictive analytics. An old, but still relevant discussion is available here.

Information Waves

Many years ago when I was a researcher in a European think-tank, I wanted to develop an automatic way to map information flows within large organisations. The basic idea was to data mine the corporate email system to discover how new information spreads through the organisation - and along the way map 'unofficial' communities of practice. I never got there because I was distracted by an offer to work in Boston on an internet project at MIT's Sloan School and the Lotus Development Corporation.

I've kept this interest in mapping information waves and I was interested to come across the recent work of Jon Kleinberg, Professor of Computer Science at Cornell University in the US. Together with a couple of assistants, he has tracked 1.6 million online news sites, including 20,000 mainstream media sites and a vast array of blogs, over the three-month period leading up to the 2008 presidential election. Analysing the 90 million articles has revealed the pulse (he calls a 'heartbeat') pattern of handoffs between blogs and mainstream media:

Here is a more detailed picture.

I wonder if a similar mapping of information flow within an organisation would lead to better analytic use?

My theory is that delivering analytics in 'harmony' with the natural 'heartbeat' of an organisation's other information flows will lead to more timely and targeted insight - resulting in better decision making. How you would achieve this is a very interesting problem. Is anyone working on a solution?

10 Signs That You Need Analytics

I live in a world where the majority of my clients have Management Information Systems (MIS) - but not Analytics. The difference? Well, MIS is (my definition):

The people, processes and technologies used by managers of an organisation to answer their questions. Management information systems differ from regular information systems because they are for analysing other information systems used in operating the organisation.

Analytics is a specialised form (or subset of) MIS that I define as:

How an organisation uses modeling (involving extensive computation) to arrive at an optimal or realistic decision based on existing data.

In other words, MIS and Analytics are both intended to support decision making and the difference between the two is that Analytics makes use of more advanced techniques to interpret data.

The above are my practical definitions that I use when creating an analytic capability in an organisation. I can already hear the shouts and objections from my former Gartner analyst colleagues and I leave further academic debate to them. These definitions work for me and maybe you if you want.

I also see that statistical analysis and data mining are closely related - but often those activities do not involve modeling to the same degree. As to Business Intelligence I see it as an umbrella term for analytics and technologies such as reporting, OLAP, data mining, business performance management, decision performance management, benchmarking, scorecarding, dashboarding, text mining, and predictive analytics.

Now that we have that out of the way, here are the top 10 signs that you need to improve your organisation's analytic capability:

1. You have to wait longer than a day for either IT or your business intelligence department to make/change a report for you.
2. Across the organisation there are more than 100 requests pending for reporting /dashboard /scorecard changes waiting for a specialist to deliver them.  
3. When you attend meetings, there are multiple numbers being quoted for the same thing - and you don't know which of them is correct. 
4. When you talk about fundamental things like transaction, account, balance or available stock - and you discover that the person you are talking to is using the same words but means something different to what you mean. 
5. You can't get an instantly understanding when glancing at a report/dashboard/scorecard and what it is telling you. 
6. The commentary is larger than the automatically generated report. 
7. The report is not generated automatically but is a handcrafted labour of love by either yourself or one of your staff, or you spend hours trying to locate the right data and then have to consolidate it manually into Excel. 
8. It takes longer than 5 minutes to view a new report. 
9. You can't access the report when and where you most need it. 
10. There are hundreds of reports available to you but you don't trust them and you spend time trying to manually validate key numbers.  

If these are the signs, then what are the solutions? That, dear readers, I leave to another blog entry ...