Before you take steps to improving your website’s search engine ranking, you should consider who the majority of your audience are and what search terms they are likely to be using, says Peter Morville, author of Ambient Findability and a speaker at this month’s Webstock 08 conference in Wellington.
Before a redesign of the US Government’s National Cancer Institute website and strategy, the site would appear high in Google lists for “cancer”, Morville says, but would not respond well to search-terms for specific cancer types, such as “breast cancer” or “lung cancer”.
Probably 95% of the people who come to the site are people or relatives of people who have been diagnosed with a specific cancer and they would most likely search under that specific term, he says. In this case, the consequences of getting less authoritative sites appearing higher on the list could be serious.
Finding the website is only the first step, Morville says. Then the user has to “find their way around your website and find your products and services despite your website”. The NCI site, for instance, has been reorganised with a home page for each type of cancer.
Navigation and retrieval of information is an old human skill and lessons can be drawn from the way people have always done it. That is by search, navigation and “wayfinding” — moving from one landmark to another until the goal is reached. Those lessons can be applied to the web environment, Morville says. Provide multiple ways of getting to the same destination, bearing in mind that not everyone has similar level of knowledge.
Another medical database was organised by the body’s systems — bones, nerves, blood, respiration and so forth. But where does a searcher for “diabetes” go? They are very unlikely to know to look under “endocrine”.
“Don’t base your information structure on your organisational chart,” he says; a visitor to a university site, for example, is unlikely to know how the faculties at that particular university are organised.
There should be both alphabetical indexes and a search engine. Some content likely to appeal to sizeable subsets of inquirers should be allowed to “bubble up” to the home page.
Design is for the future as well as the present, Morville says, a future that will have an increasing flood of information, where information will be ambient — all around you — and so will information-acquiring devices, with attendant risks to privacy.
The flow is already “a firehose aimed at a teacup” and people will increasingly learn to pull specific information through selective devices rather than have it pushed to them. The range at www.ambientdevices.org perhaps shows the direction; a device accesses a range of data through the internet and may indicate a digest of it through a few figures, a graph or even simply by changing colour.
Ambient Devices makes the Ambient Orb “a frosted-glass ball that glows different colours to display real time stock market trends, traffic congestion, pollen forecasts, or any other ambient information channel.”
Metadata, previously an interest of specialists, “has become sexy”, says Morville; everyone is designing taxonomies to classify the data objects they own or allowing their users to “tag” objects subjectively with a category they think is appropriate, creating so-called “folksonomies”.
But whatever intuitive searchable information structures are designed, he says, “decades from now, we’ll still be entering keywords into a box and hitting ‘go’.”