The web is not TV: Avatars break the web contract unless used very carefully.
Why is the web different from TV in this case? The TV contract of talking heads, sound, voice is signed when the ON button is pressed. Not so with the Web. What is the Web contract?
– No talking unless I say so – you should be able to articulate your idea without sound initially
– I start the interaction, you can prompt me with animation and some sound effects or an introductory story, but not a small talking person.
– Avatars are typically Cheesy – which is the term used when something was used in the past strictly because it had wow factor, not inherent usability value. Having a greenscreen character walk or beam onto a portion of the site restricts site layout and simply does not have the payoff that a dedicated video window has. If the whole site is video, that fine, but combining a CSS-typ layout with an Avatar is a kludge trying to bridge two different media.
What is the Avatar’s contribution? Are they interactive with a branching script line based on interactivity? Are they familiar from other media in the marketing or content (ie. Do they have star power or tie-in) ? Are they part of the product itself?
An avatar has to have a very good reason to be included on a site, and has to be executed as a focus of attention and not a Christmas decoration in the corner of the site. People expect the Avatar to be different that TV – it has to respond or be interactive, moreso that Clippy!
[image from whirlingturban.com]