Last week I had the pleasure of presenting on the topic of UIA and how it drives a 'return on information' - the new ROI - during a webinar with Matt Brown from Forrester Research. You can view the webinar on the KMWorld website.

We didn't have time to get to questions so I thought I would tackle them online.

Q: When we were more of a manufacturing marketplace, measuring results and performance was relatively straightforward. It is much more difficult nowadays as service and information providers to measure and grade those same things - not enough info/too much?

A: I think understanding the customer and how they view your products and services has always been important; it has simply been hard to analyze until the internet enabled the incredibly rapid creation of digital content. Social networks, blogs, wikis, etc, all offer a non-stop stream of feedback and competitive insight. No company can afford to ignore it. Technology like sentiment analysis can be used to comb through the data and understand important trends.

Q: How realistic do you think 2-3 month timeframes are to deploy a major ECM application?

A: It is absolutely realistic for a first phase, with good prioritization and control of the variable parts of the project scope. Some of the biggest drivers of effort are templates, schemas and workflows. If you have to author lots of those the risk of delay is high. If you have to customize them the risk is lower.

Q: One of my problems is that information is not published at all, or it is published only on paper. How can a UIA strategy address?

A: Optical character recognition (OCR) is one route. Entity extraction and pattern matching can be used to link up paper documents with digital ones.

Q: If our data silos are using different taxonomy and metadata schema, how much preparation is required?

A: Hopefully your UIA platform is 'schema dynamic' meaning that it creates the schema as it ingests data from each silo. This effectively takes the "T" out of Extract, Transform & Load (ETL), so long as you don't have to flatten too much of the data. (If you do, you'll lose some of the important relationships.) Attivio AIE is schema dynamic; by default it can search across all sources, or within any source, or can present you with a list of sources and fields available to you. Taxonomies can be applied to their individual sources. If you want to normalize them then a mapping effort is required.

I'd like to thank everyone who attended and invite them to This e-mail address is being protected from spambots. You need JavaScript enabled to view it with any further questions or feedback. Thanks also very much to Matt Brown for a terrific presentation, and everyone at KMWorld who hosted the webinar!

TrackBack URI for this entry

Comments (0)