With more than 200 million web sites in cyberspace these days, what makes the difference between the good ones and bad ones? That’s what Stanford University has sought to find out through a multi-year research project.
The Stanford Persuasive Technology Lab conducted research with a cross-section of more than 4,500 web consumers over a three-year period. Distilling the mountain of information gleaned from this sample, the Lab issued its Stanford Guidelines for Web Credibility. It boils the research findings down to ten basic guidelines that, if followed, mean that a web site will be viewed as credible and authoritative.
In reviewing the Stanford guidelines, some of them stand out to me as ones that are too often missed by web developers. In particular:
Show there’s a “legitimate” organization behind the web site. Listing a physical address – not a P.O. Box – is one way to do this.
Make it easy to verify the accuracy of information. Third-party citations and data references, along with links to other respected web sites, are ways to accomplish this.
Make sure the site is “intuitive” and easy to navigate. People’s tolerance level for a web site that doesn’t follow a clear, logical thought path is low.
Update site content often. People put less credibility in sites that look like they’re informationally static or stale.
And especially important: Stanford’s research found that content errors should be fixed, no matter how inconsequential they might seem. It turns out that broken links, bad spelling, poor punctuation and other typos negatively affect the credibility of web sites more strongly than many other factors — and yet they’re among the easiest items to fix.
When you look at the overall web credibility guidelines from Stanford, they aren’t particularly challenging – and they shouldn’t be hard to put into practice.
But when you consider how lame many web sites actually are, it’s another reminder of how – in web design as in so much else in business and government – “best practices” too often fall victim to expediency or just plain slipshod execution.