TCANZ 2010 day 1 – Metadata: Key to a successful intranet
I’m attending the TCANZ Conference 2010 in Wellington, New Zealand. Ann Rockley gave the keynote address this morning: “Metadata: Key to a successful intranet”. These are the notes I took during the session. All credit goes to Ann. Any mistakes are mine.
Ann started with a general overview of the problems an intranet may encounter when it starts growing, gaining lots of content and possibly suffering from haphazard organisation. Then she showed how you can reduce the problem by good use of metadata.
Your intranet is only as good as your employees’ ability to find information in it. The key to retrieving information is metadata.
Metadata is “information about information” (rather than “data about data”).
These are some of the uses of metadata that Anne talked about:
- Similes. People won’t find information if they search for a simile, a word that’s not in the document but means the same thing as a word that is in the document. To solve this problem, you can add the similes in the metadata.
- Categories. You can also use metadata to categorise information. For example, you might tag information for specific user groups, such as business analysts.
- Tracking. Metadata can mark information that is in a particular state, such as its status in a workflow (draft versus approved, or translated versus awaiting translation).
Ann talked about taxonomy, which is the process of organising information. Metadata is one of the tools you use in your taxonomy. It is important that you build your taxonomy with your users in mind, because they’re the ones who are going to use it. You should also build it in such a way that it can grow. Ann gave us some interesting insights into the limitations of the Dewey Decimal System.
You also probably need a controlled vocabulary, to make sure that everyone in the organisation is talking about the same thing. This is especially important if the organisation has teams of people in various geographic locations.
A taxonomy consists of:
- File structures.
- Navigation structures.
- A controlled vocabulary – A list of agreed terms. Ideally, only certain people can add terms to the vocabulary. This means that people know what to call an item when they add it to the intranet. People also know what to call it when they’re searching for it.
- The relationships between items of information – Synonyms; broad versus narrow terms (furniture and bed); preferred versus variant terms (colour and color); etc.
- A thesaurus.
Ann led us through the steps to take when creating a taxonomy. The most important things to concentrate on are the goal of your taxonomy, and the people who will use it. Don’t focus on the technology in the early design phase. She gave us useful guidelines on identifying the scope of the taxonomy. We looked at the details of designing and finding hierarchies, facets, keywords, etc, in the information that your taxonomy will serve.
A good tip: If your information is on a website, use web analytic software to find out what people are searching for, and add those terms to your keywords.
Ann wrapped up by telling us about the books she has published and those in the pipeline. She also mentioned the Intelligent Content 2011 conference, taking place in Palm Springs in February 2011.
One of the questions at the end of the session was, “What are your strategies for getting people to agree on the terms used in the controlled vocabulary?” Ann’s answer:
- One great way is a “card sort”. Put the terms on index cards and see how people use the cards when they try to find information.
- Another good way is to see what people are actually searching for, on the existing intranet.
- A good way to defuse the situation is to use preferred terms and variant terms, and let people know that they can still use the alternative terms on the intranet.
Always remember, it’s not about the people designing the taxonomy. It’s about the users.
Thank you Ann for a detailed and considered walk through this aspect of information design!