How to Create a Know-it-all Company

In the current cost-focused climate, plenty of organizations are calling on competitive intelligence to help reach educated strategic decisions.

1 2 3 4 Page 2
Page 2 of 4

If a sales rep out in the field is already recording what doctors they see and the products that those doctors ordered, it is far better to add another field to record those doctors' individual responses to Glaxo's initiatives in relation to asthma than to introduce an entire new system to capture campaign intelligence. "If they can do it as part of their normal call record activity, then it's more likely to get done," Ooi says.

He warns that using systems instead of process as your starting point exposes your organization to the danger that the money it invests in CI technologies could be wasted.

"[If] someone implements a knowledge management system or a huge database or some sort of online forum to collect data, and it's done without the processes around it that link it to someone's normal job responsibility - we've had the same experience - no one enters information into it. They can't see what they get out of it, or if they can see but it's just not part of their normal work routine, they kind of forget its value. So you often don't see the value coming out of that investment at all," he says.

The Human Element

However important the technology, CI is still mainly about people, and about planning, says Professor Daniel McMichael, group leader for business intelligence CMIS at CSIRO's division of Mathematical and Information Sciences. Over the past 30 years technology has provided ways of reducing the cost and computerizing and improving the quality of low-end activity, according to McMichael. Once we didn't pay typists very much, for instance. Now we all do our own typing and word processing, and while on the whole we get much better results than ever we did when we had typists, we spend more time doing that work. "So that's good, but if you can sell to a company or any organization the possibility of replacing or better utilizing much more expensive talent, then they're going to like it," McMichael says.

One way the CSIRO is trying to do just that is by working with Boeing in the area of "source fusion", looking for techniques to synthesize useful knowledge from collections of data, as a way to move CI up the value chain. People know how to do searches - you put in a query and end up with a whole series of documents. Now the CSIRO is looking for ways to help turn searches - probing both the Internet and relevant databases - into investigations, by providing technology that can extract the relevant items in the relevant documents and then begin to assemble them into a structure that is "halfway towards a document". "What you then need is the technology to do all that stuff. And a lot of it is based on search, but a lot of it is based on language understanding, so that you actually get semantics based on meaning out of a document and sections of documents. So you can apply that technology not just in CI but in intelligence analysis for Defence, counter-terrorism, all this kind of thing using exactly the same technology."

In the meantime, McMichael recommends organizations take a look at the CSIRO's P@noptic, a new search engine for corporate and government intranets that he says makes information more accessible, improves information flow to clients and significantly increases efficiency. But McMichael says the main lessons to have come from the CSIRO's extensive experience in CI are that you should never expect the machine to do too much, and that tools must be well integrated to be truly effective. "You want to have human involvement at all stages," he says. "You want to be able to say: 'Okay, I need to get involved in this if the machine isn't working very well.' So you can use whatever technology you're given, in the best way for you, and where there's an automatic solution you can switch it out.

"And the other thing that I think is useful to be able to do, is to be able to plan what you're doing around an investigation," he says. "So instead of just having a whole series of different tools, relatively independent of each other, the tool enables integration of a number of technologies. I don't know about you, but I spend a large amount of my time transferring information from one computer tool to another computer tool, and if that process can be automated so that there's effective integration, then that's a massive saving."

McMichael says currently the best option for integration is a workflow solution, although since workflow is really designed to "shovel information around the organization to different people working on a single job", this is far from the ideal. People are also at the centre of considerations of the need for CI, says Helen Price, market research and planning manager (Oceania) for DHL Worldwide, and those needs must be the central consideration when choosing tools. In DHL Worldwide, management tends to want to do ad hoc searches and ongoing tracking, while employees mainly need the sort of general information Price makes available on a basic intranet.

Until Price came on board, DHL Worldwide provided a so-called "competitive toolkit" to help managers do CI, but there was no structured approach to the organization's competitive intelligence efforts. Price has been working to change all that, introducing a formalized program at the end of 2002.

1 2 3 4 Page 2
Page 2 of 4
Download CIO's Roadmap Report: Data and analytics at scale