Is google happy indexing tabular data?

[downhiller80]

I've just launched a new website, that accumulates (with permission) results for my key sport. This is a very unique idea, that no-one's done before in this niche, and it's proving amazingly popular given the relatively small target audience!

I went live on Monday (4 days ago) and submitted a sitemap to google at the time, containing about 4000 URLs.

I was impressed that within a day google had already indexed 150 pages (site:domain.co.uk search). The next day it had dropped to 100. Then 50, now 15 :(

Is this nothing to worry about since the site is so new, and just google shuffling things about, and in a few weeks I can probably expect to have most of the 4000 pages indexed? Or is it likely a problem with my formulaic pages?

Each page is very low on non-tabular content. They have the athlete's name at the top, and then a table of all of their results, showing the following columns, half of which are purely numerical. Is google happy indexing this sort of page? If those 4,000 names get indexed it'll bring in a fairly sizable chunk of traffic - well, sizable by this site's standards!

Columns:
date / event / venue / category / sponsor / position / entrants / beat% / best time

cheers!

[edited by: tedster at 12:48 am (utc) on Jan. 22, 2010]
[edit reason] removed some specifics [/edit]

[tedster]

By "tabular data", I assume you mean the information is displayed in tables. Yes, Google definitely will index that. I think your idea about being too new is exactly the issue. Your site probably needs more traffic and more backlinks before Google will keep more of your URLs in the index.

And by the way, don't believe the site: operator stats. Better to watch your logs and see which URLs Googe sends traffic to. That's going to be more complete and dependable as a metric. And if you want to check on whether any particular URL is indexed, just enter it directly into the search box and see if you get a result or not.

[downhiller80]

Cheers ted. And yes, I do mean in tables. I have clickable column headers to resort the data on every column too, I've put rel="nofollow" on these so that google doesn't try to reindex the same pages just in a different order - seemed a nearer/better solution than using the parameters section in google webmaster tools - do you agree?

[tedster]

Yes, it's definitely a good idea to keep different "sorts" out of the Google index when they generate different URLs.

Not that the parameters exclusion in Webmaster Tools is bad as a back-up solution - or an insurance policy, for that matter. But it's always best to assume first responsibility on your own server whenever you can. The parameters exclusion tool is very useful for situations where other approaches are just not practical because of the website's complex infrastructure.

[downhiller80]

Thanks again tedster. May I ask one more question? Should be a new topic really, but since I have you in conversation here seems pointless. Wondering what you think of my URL structure. For those 4000 pages I'm going with

www.domain.com/athlete8728/firstname-surname[-modifier]/

Obviously the number is an example and changes for each page. The modifier is if there's two athletes with the same name, I have a very short note so people know which is which.

Is that as google-friendly as I can make it? Or will the large ID numbers (up to 5 digits) put google off, and I should drop the "/athlere2281/" bit and just go with www.domain.com/firstname-surname[-modifier]/ and accept that my links will break if I need to change someone's name/modifier at any point?

Thanks again!

[tedster]

As long as the keyword is in the URL, and each URL is uniquely aimed at one bit of content (no canonical URL issues), then your URL structure has done its job. So I'd say you're fine.