When I look at many newspaper websites, I just can’t help but wonder why some stories are in a featured homepage slot and, even more so, why they remain there for so long.

I recognize that every market has it’s own issues and what is a popular story in say North Jersey, might not work in Ann Arbor, Michigan. So while I sometimes wonder if “Gardening to Extremes” deserves top billing, I will ultimately bow to the taste judgment of local editors. But what I will question is why that and many other stories keep that top billing throughout the day?

Looking at a variety of sites I see a couple of patterns at work: There are sites that keeps their top stories in place until there is breaking news. There are the sites that rotate content based on the time of day. There are the sites that don’t change anything and even breaking news is only treated as an update. And then there are those that appear to take an approach that I think all sites should emulate. The producers and editors at these news sites let their audience guide them.

It make little to no sense to me to keep a story in one of your prime spots if it’s not drawing an audience. Honestly, I don’t care if it was the top of 1A and you think it’s a hoot or the most important story “evah”, if it’s not bringing them in, then what are you achieving?

How do you deal with this at your newspaper? Do you pay attention to your analytics and let the numbers be your guide? I know that previous posts both here and elsewhere have caused readers to bristle over the use of analytics. But come on, can you argue that knowing what is or isn’t working or your site is really a bad thing? Newspapers spend a lot of money for analytics. Omniture, one of those more popular services starts at $20,000 and easily climbs from there. If all you are using it to find out what was popular before the morning meeting or on the previous day, then I would argue that you’re wasting good money.

So back to that nautical talk. I think you should be able to change the direction your site is going as frequently and as often as you find necessary, that you’re always offering your best content (whatever that may be). My suggestion is that you come to an agreement on a threshold for keeping or moving content based whichever metric you want use, and then, by golly, tack to the starboard or tack to the port, but turn your ship.

As a reader, it is great that I can go online and learn that the mayor of my town had an estimated 2006 salary of $7,000 or that an elementary school music teacher can evidently make $100,183. But the same newspaper site that can tell me all that, can’t tell me where the mayor’s office is, how to contact him or how to join the local Parent-Teacher Association (in case I wanted to find out more about that teacher’s salary).

Newspapers want people to use their sites and to keep returning to them. They discuss ways to make them “sticky,” more attractive and more fun. But I think in failing to provide readers and site visitors with this basic level of information they’re missing out on an obvious opportunity. And even worse, this forces readers to leave these sites, when more then likely they’d be content to stay.

OK, so what is this basic level of data that I am talking about? Well, I think it’s as simple as providing the who, what, where and when. And then you can layer on the additional pieces. For example tell me:

• the name of my mayor
• how long his/her term is?
• their biography and/or resume
• where is their office
• their phone/fax number
• their email address
• their web address
• How to schedule an appointment with them? For example, are they available before council meetings?

From that basic information, you can then provide links to the mayor’s salary/pension information, council meeting agendas and resolutions, meeting minutes, speeches as well as stories, photo galleries and videos that they have appeared in on your site. And you don’t have to stop with politicians and politics, this approach can be applied to every aspect of your communities, including how to libraries and even building departments — think of all the ways people interact in your coverage area. It is almost local search for civic life or Yellow Pages on steroids.

The benefits of this type of approach are multi-fold – you can bring some of that data you have already organized out of its data ghetto; you make your site the single, definitive source for local news AND information; and you can aggregate all the information that your reporters should already have and is likely scattered throughout your newsroom.

I think you’ll find that this approach will score more points with readers and help increase traffic to your site. You don’t have to do it all at once, start small, pick a few communities and/or organizations and go from there. And if you have another way of approach this, I’d love to hear about it.

But there is a problem with many of these sites. In addition to being what Matt Waite so entertainingly calls “Data Ghettos”, no one has figured out how to monetize them. As a result, the sites are not really designed with the readers or end-user in mind, and little to no thought goes into their longevity. What does go into these sites is a disproportionate amount of resources — money, staff and equipment.

What makes this situation even sadder is that many of these data sites, regardless of extensive their information is, are not even used by the reporters and editors at their own newspapers. I have colleagues, for example, at Gannett newspapers who say that they don’t go to their DataUniverse sites and will rather make a phone call or look up the information elsewhere.

So lets review, money and manpower is being invested to create and maintain these sites; advertisers are not attracted to them and are not purchasing ads on them; they are not used by their own staffs; and they are not even drawing a lot of traffic. Compete.com claims DataUniverse.com drew only 3,050 uniques for the month of November. Even if you disregard Compete’s “rough estimate”, one has to question the continued viability of these sites.

Now don’t get me wrong, I love data. I believe that newsrooms typically have way more data then they realize and that harnessing and organizing that data is key to newspapers being able to truly compete in the world of the 24-hour news cycle. But is this the best we can do? Really?

I think that we need to stop treating these sites simply as editorial sites. They are not some holy cow that can’t be touched by the revenue side of the business.

I think we need to approach the creation of data sites much as if they were verticals (you know, like our real estate or automotive sites). Yes there needs to be a place for content and discussion of user interface, but there needs to be an equal discussion of the revenue opportunities and the sites marketability.

Given the state of revenue at many newspapers, I don’t think we can continue investing in online products that stroke egos, provide little continued service to readers and brings in little to no additional revenue.