Cheat Sheet: Business intelligence

Update: Another oxymoron?

'Business intelligence'? Is this just another oxymoron, like police intelligence or military intelligence?
Don't be so harsh, though the basic principles of all three are certainly the same - gathering information and making informed, critical decisions.

OK. So what kind of information?
Almost anything really - any data that a business creates and which could give it a strategic insight into what it is doing well, what it is doing badly and more importantly what it could do better. One area where this is proving useful is in retail.

Tell me more…
Let me start with a simple example. If you run a small corner shop you may think you need to order a certain amount of daily newspapers and monthly magazines. You might also buy milk, fizzy drinks and beers for your cooler cabinet and some crisps, sweets and basic food such as bread and eggs for the shelves.

Well, if you know there are major events coming up, such as the World Cup, you might think to order in some extra crisps and beers for the couch potatoes out there. Likewise, you'll probably remember to buy in some extra boxes of chocolates around Valentine's Day. If you don't know these events are happening you might miss out on some extra sales, and disappoint potential customers.

Sounds pretty straightforward to me. Do you really need tech to help you remember when big events are coming up?
BI software does a lot more than merely remind supermarkets to buy more turkeys at Christmas and disposable barbecues in summer. Business intelligence software will cross-reference multiple factors - from seasonal change in demand for goods, to the relative cost of shelf or rack space and availability of replenishment - and it will enable retailers to make decisions based on all that data.

So what else can BI do besides control stocking?
Well, there's no point having full shelves of carefully priced goods, stocked 'just in time', if you haven't spotted a need for higher levels of staffing, exacerbated by increased illness and absenteeism during winter, which will impact your busiest time - Christmas.

Surely retail isn't the only area where BI can help though…
Good point - in theory BI can help any business that creates operational data to gain an advantage from understanding that data better, most typically by getting better insight into user or customer behaviour.

It's not just about dealing with customers though: businesses can also use BI to be inward-looking, monitoring their internal workforce for everything from anomalous data, flagging potential fraudulent activity, to patterns of churn in key departments.

Give me some examples of these other uses of BI then.
Well global bank HSBC is using analytics technology to spot fraudulent activity in payment-card and online transactions.

Meanwhile, B&Q is using data mining to root out internal fraud and reduce stock loss and insurance group RSA is using BI to further help its contact centre staff improve customer service.

It sounds like this BI stuff is pretty limitless…
You're starting to get the picture but we should also add that any technology is only of use when used effectively. Simply buying business intelligence software won't make your business more intelligent - it's only as good as the people who know how to react to the data. It won't necessarily order your turkeys for you, it certainly won't interview a potentially fraudulent employee and it won't hire the best staff.

What can businesses do to use data better then?
A lot of businesses merely produce reports using the data at their disposal without really examining it to work out how to improve different areas of the business. If they take advantage of more advanced technologies such as predictive modelling, text analytics and performance management, they could make this data much more valuable.

In order to do this, experts suggest the business side of organisations should become more active in communicating what they want to achieve with BI and not leave this to the IT department. The current economic situation is likely to make organisations look more closely at their BI technology to work how they can exploit their data more effectively.

Businesses must be all over BI then…
Actually, it seems that although BI technology has developed in leaps and bounds in terms of what it can do over the past few years, businesses are still not making best use of it to improve their business.

Gartner suggests that even by 2012, more than a third of the top 5,000 global companies will fail to make "insightful decisions about significant changes in their business and markets". The analyst house suggests this will be down in part to a lack of investment in information infrastructure and business tools - or business intelligence technology.

That said, that leaves a lot of companies that are using BI. Business intelligence also regularly appears on lists of CIOs' spending priorities and Gartner picked it as one of strategic technologies for this year.

I guess the tech industry is getting pretty excited about all of this…
And you'd be right - IBM, Microsoft, Oracle and SAP all have their finger in the BI pie having made some significant acquisitions. In recent years IBM has shelled out around $5bn for Cognos, SAP acquired Business Objects and Oracle has brought Hyperion into its fold.

This means BI is increasingly being incorporated in the software stacks that these companies offer which would obviously suit businesses who prefer to stick with a few large suppliers. There's also potential for people like Google and HP to become more active in the BI area.

There are numerous other BI players outside the traditional tech powerhouses, notably SAS and SPSS who cater for vertical industries - something that some experts say could be a growth area.

Well if all of those guys are involved it must be a pretty big deal. What's getting people excited with BI at the moment?
The proliferation of data available on the web is something that is becoming increasingly discussed. Companies have more access to data outside their company that they can incorporate in their BI work. Text analytics in particular could become more prominent for processing information from web documents and content.

Gartner suggests we will see more businesses making decisions based on BI technology combined with social software. By using data generated by relevant communities, businesses could use social and business networks to make more informed decisions.

Another area of potential interest is that BI technology itself could move online with "information as a service" becoming more prevalent.

I have to say I'm a lot more interested now than when we started our little chat.
I'm glad to hear it. Just remember that although business intelligence tends to put us in mind of large software companies providing huge database-driven software apps, this is first and foremost a way of working rather than a specific technology - gathering more intelligence on what we do.

Well if you put it like that, how can I resist?
How indeed…

Editor's Note: Will Sturgeon also contributed to this story.