Nielsen continues to miss the point with its decision to replace page views with “time spent” as a key metric in deciding which sites are most successful. No single metric or pair of metrics can provide a complete picture of how well a site performs. Instead, I suggest our industry learn a lesson from the big credit card and mortgage companies. Create something akin to a credit score. Call it a traffic score.

The traffic score should combine many metrics into a single weighted number. Include page views, unique users, repeat users, time spent, and page views per visit to start. For example, if all five of those were weighted equally then each would contribute to 20 percent of any Web site’s traffic score.

Since Nielsen believes unique users and time spent are the most important indicators of success, then maybe it would weight those at a higher percentage of the score than other factors. The traffic score provides a more comprehensive story.

The idea of a traffic score isn’t a hard one to grasp, so I won’t belabor the point. Except I do want to point out the effect of focusing on page views was too often to encourage bad site design and disregard new technology such as AJAX. The effect of focusing on time spent could mean making a page more confusing actually leads Nielsen to believe it’s more successful. Focusing strictly on unique users lets people focus on driving up traffic from search engines and ignore creating a loyal user base. I could go and on, but that's enough to make the point that any single metric provides a partial picture.

Nielsen, ComScore and others should create a traffic score. Perhaps Omniture and Web Trends will lead the way by first creating a traffic score within their systems.

RELATED: Tabbed browsing not counted by Nielsen as time spent