When tags work and when they don’t
Categories: Web 2.0
Tags: amazon, folksonomies, folksonomy, librarything, tagging, tags, thingology
I’ve talked before about LibraryThing as one of the best examples of advanced use of tagging tecniques.
LibraryThing (11M of books) lets users catalogue and tag books. Subject headings are shown side by side with people assigned tags. You can browse using a top down approach or fly horizontally leveraging tags. A bonus is the possibility to create synonyms among new and existing tags.
Thingology, their blog, is very often a good source of inspiration for tagging related things.
In my opinion, the single most important question regarding the use of tagging is when using it at all: why in some cases (in Del.icio.us and LibraryThing for example) people flock while other times they simply tends to ignore the possibility to describe objects through medatada?
The Del.icio.us Lesson is probably the most fertile ground to formulate an answer (see also Why people tag) but providing precise guidelines to find out the best opportunities for tagging is still not easy.
Take Amazon: they introduced tags at the end of 2005 on more than 60 million registered customers and with one of the highest traffic on the web. A sure bet! Sorry, but users never got it. 60 million users translated only into little more that 1M tags.
Take LibraryThing now: tags unveiled circa at the same time, a lot less users and traffic but more than ten times that number of tags (13M).
Why and how can we abstract these reasons to other e-commerce sites? Thingology gives the answer with their post When tags work and when they don’t: Amazon and LibraryThing.
I suggest to go there and read but here is a quick summary of the conclusions:
- Amazon could have done more to get their user tagging. So user experience and user support are not optional for the success of a tagging initiative
- You need a good number of tags on each item. More tags mean quality and noise reduction and you can do a lot more when the data is so rich
- Tags are useful only when you have a lot of things to remember. Heavy users bring more tags
- Tagging has to be an engaging, funny and social experience inside the system. People don’t tag for the sake of it
- Users want to own and being really involved by the application. Provide them with enough control or you will loose them
- Opinion or ego-nature tags (nice, ugly, toread, etc) are dangerous in a low-numbers environment. One user alone shouldn’t be able to completely influence the system
And finally some precious advices:
* Figure out why your customers would want to tag your stuff. Don’t fool yourself.
* Make tagging as easy as possible. (Amazon’s are quite easy to add, although registration is a pain.)
* Understand that commercial tagging can turn people off. Avoid crass commercialism. Respect your taggers—these people are helping you out!
* Make taggers feel like it’s “their” thing. Encourage users to give out their tag URLs—people love to show off—and let them export their tags any way they want.
* Keep tagging social. Stop selling and start connecting. If you connect people up right, the selling will follow. Think Tupperware!
* Consider whether a non-commerce site has the data you need. Back when LibraryThing had a million tags, Amazon could have bought our data for the price of a cup of coffee. Now, that we’re big and important and have three employees, that’ll be THREE cups of coffee, buster!
In the same category:

February 24th, 2007 at 5:59 pm
Hey, thanks for the blog mention. Your “Why people tag” is dead-on. Just a quick note, we’re at almost 11M books now, not three. That’s actually telling in a negative way. Most books are not tagged, but tagged books have a few tags. I can’t remember the numbers, but I think it’s something like 3 tags/book when there’s a tag.
Thanks! Tim