You May Have Metadata, but Is It Actionable Knowledge?
I hope all your content has metadata. It probably doesn’t. Those pesky end users can store a document every which way. Then those same end users can’t find it. And they spend an average of 4.5 hours per week per person looking for it. It’s a vicious circle.
Do you think your organization values data as an asset? Do you think your organization has a clear data strategy, implements it, and staff embrace it? I would think very few of you would answer “yes” to the second question. With the continual influx of unstructured content, which represents over 80 percent of your data, it might be a good idea to start thinking about your data strategy. Legal, compliance, and records management teams will greatly appreciate it.
Now the tough question. Do you think your metadata represents insight, intelligence or knowledge? The ultimate goal is to give people not only the right information, but information distilled from a variety of distinct content, making available useable knowledge.
We can keep waiting for artificial intelligence (AI) and associated technologies to, maybe, solve our problems. But what do we do about things now? For those blog purists, I’m now going to offer some sales type information – I’m not in sales, so it won’t get too salesy – and I ask you to take a moment to understand how metadata can reflect knowledge.
We all know how hard it is for end users to consistently apply metadata – even when provided with a drop-down list, they will overwhelmingly select the first option, rendering the metadata useless and just plain wrong. And only about 20 percent of organizations employ effective metadata and classification of their data. No wonder enterprise search is often a dismal failure.
Many auto-classification products on the market today require complex rules to be generated, often involving search syntax. Some even require a document training set for every term to be processed, or demand the use of Boolean expressions to create rules – not a common skill. These techniques create a high initial cost, in terms of both time and qualified staff needed.
Products employing linguistic techniques will not perform consistently across different vertical markets, have a tendency not to scale, and can perform poorly. Think about it, the grammatical style of a legal contract or patent application is very different to that of a news article or a typical web page. The inability to identify intelligent metadata from within content impacts the entire lifecycle management of the asset.
Generating knowledge, or intelligent metadata, requires an adaptive technology that
- Generates multi-term conceptual metadata at source
- Is not based on keywords, proximity, word counts or algorithms that cannot be changed
- Captures very specific criteria for business applications using the metadata
The end result is a rich set of intelligent metadata that reflects the unique terminology and vocabulary of an organization, and so makes it valuable for search, records management, compliance, legal and eDiscovery, and data privacy. Data is auto-classified at source, at the time of ingestion or creation, and the knowledge metadata generated becomes the basis for executing workflows to automate business processes.
And what, pray tell, delivers this magic? The conceptClassifier platform includes all you need – generation of knowledge metadata, auto-classification, and taxonomy tools to manage the metadata. It works out of the box, classifies content in any repository from any environment, and is designed for subject matter experts not just the IT team, but we let them use it too. What’s not to like?
In any case, if you aren’t generating knowledge metadata, you are missing a key component that can radically improve your data strategy. You are wasting money, losing productivity, blocking innovation, are more vulnerable to data breaches and noncompliance, and will face increasing disruption and unknown risks. Sounds horrible, doesn’t it?
Let’s take a look at our first two questions again. Do you think your organization values data as an asset? Do you think your organization has a clear data strategy, implements it, and staff embrace it? It would be great to be able to answer a resounding “yes” to both questions. Now you know you can. Hey, if you made it this far, thanks for reading.