What A Cute Bunny: Taxonomy as Liberator

I spent this past week testing a taxonomy as part of a digital asset management project we are currently working on. One of the test scenarios involved giving art taggers a series of images and asking them to code them using the taxonomy we had developed.

Taggers see taxonomy as a blessing and a curse. On the one hand controlled vocabularies are a tagger’s dream; a nice list of consistent terms that alleviate the problems of free-tagging (e.g. five variations on the same term, plural vs. singular, spelling mistakes, etc.) However, these same vocabularies quickly become a tagger’s nightmare when they perceive the values to overlap or be ambiguous – especially if you are used to only being able to select one value from the list.

Have a look at the following image

Easter Bunny

How would you describe it?

Style= Cute? Silly?? Whimsical???

(note: extra ?s  added to denote level of rising anxiety in tagger)

Taggers are a lot like us (us being taxonomists): we like it when things fit neatly into categories.  In this case however, the beauty is that this faceted taxonomy was specifically designed to accommodate the fact that most images don’t fit into one category. Watching the nervousness and anxiety wash away when we explained to taggers that they could use as many taxonomy values as were appropriate to tag an image was the highlight of the week. Instantly the taxonomy transformed before their eyes from a categorical prison into a structure of possibilities.

This is not say to that ambiguous values and overlap are desirable – we definitely want to minimize that. Just that the flexibility in applying values makes it less critical that they all be strictly mutually exclusive. We can be silly, cute, and whimsical, or sometimes just plain silly. The important thing is that options are there.

Audience Modeling

Ever since Polish biologist Jastrzębowski coined the term “ergonomics” in 1857, we have been trying to decipher the tricky relationship between machine and human. Regardless of whether you’re designing front-end interface functionality or crafting an information architecture that serves as the clothes hanger for your content, user-centered design is undeniably a major player in achieving results. It’s certainly a crucial consideration in every taxonomy project. Project stakeholders often envision the ideal taxonomy as being “all things to all people”. It’s a wonderful idea in theory, but the resulting structure would be a chaotic mishmash that in practice fails to meet most user needs. Part of the solution is in the art—and science—of understanding and categorizing your audience.

Creating usable systems requires user research in order to understand audience goals, needs, and capabilities. This process results in a host of raw data, but the data is only valuable in its effective application to system design. Quantitative data can be measured, charted, and graphed, but that only tells one part of your audience’s story. Popularized as an interaction design method by Alan Cooper, personas are a decision-making and communication tool that can organize the numbers in your data, but also describe patterns of why actual users behave in their particular ways. This qualitative information is the meat of reliable personas. Continue reading

Evangelism Marketing

Testing and validating a taxonomy can go many ways.  With a little luck and some hard work, usually it goes pretty well, you watch users click through the structure, find the right terms, and you go home feeling like everything’s in its right place. 

There are always the nightmare scenarios, the tester who can’t find anything and randomly clicks through the taxonomy as though they were sight seeing on a Sunday drive in the country, the tester who looks in the same category for everything… don’t ask me why but it’s always accessories, the tester who freely admits volunteering for this session to escape an insane co worker, if only for 45 minutes…  But I digress.

This last testing experience was by far and away the most rewarding I have ever done. Not because the taxonomy was perfect…it wasn’t.  Not because the testers were brilliant enough to intuit our mistakes… they weren’t, but because almost everyone left the sessions excited about taxonomy in general, and thought it would make their life, at least the part that had to do with information management, better.

I know, its sounds crazy, you don’t expect a positive response when you tell someone that from now on when they create a document they are going to have to use this elaborate structure in front of them to tag it…but they were. Everything from “Its about time”  to ” this is going to make finding things later a lot easier”.  Now certainly a large part of this positive reaction had to do with the fact that the overall structure and vocabulary of the taxonomy really resonated with them, but an even bigger part of this positive reaction had to do with the sense of participation in the overall project the testers felt when truly engaged and asked for feedback. 

What I am really trying to get at here is that taxonomy testing can be about more than just getting confirmation that you are great taxonomist, that people can find what they are looking for and that you will in fact most likely work again. 

Socialization of a taxonomy project through user testing can help to build project momentum and a sense of ownership of the taxonomy well before you ever actually force someone to tag a document with it.  So test early and test often, and try to ensure that when eventual users walk out of the room they feel they have contributed something meaningful to the taxonomy development process.

It may be a stretch to think your taxonomy project will be the subject of water cooler conversation for weeks to come, but if you can get your testers excited enough about taxonomy to market it for you then you are one step closer to a succesful project. 

 Let your users be your evangelists

The Pain and Gain of Taxonomy User Testing

As a taxonomy consultant, I always recommend (rather, urge with great gravitas) to my clients that they reserve some time and budget for adequate user testing. As they say, the proof is in the pudding: there’s nothing better than quantitative data to tell you whether you’ve built a structure that really resonates with your core audiences and facilitates their tasks. Creating a taxonomy without testing is putting a lot of faith in guessing – albeit, usually pretty good guessing, based on industry experience and knowledge of best practices if you have a good taxonomist.

Having done user testing on taxonomies I’ve built a few times, I compare the feeling to what I imagine it’s like a being an actor or actress watching yourself in a film.

Continue reading