It’s the biggest, if not only reason we’ve long recommended against heavy use of Flash. Now Google has added a reason for you to care about the speed of your web site: faster web sites will get higher rankings.

This troubles me.

While it’s obvious that faster is better, making speed a component of the formula that decides who matters most favors the better-capitalized business over the smaller one, further diluting the democratization of the web. In other words, if you have a better server with better bandwidth, this gives you an advantage.

But that advantage has nothing to do with the quality, accuracy, or popularity of your content

The place this is most concerning is when a web site is hosted on a shared server and/or using a shared IP address. Somewhere, servers have built into their operating parameters a “maximum connections” rule. If you’re buying the typical $10/month hosting plan, therefore aren’t a machine-level administrator, and are sharing space with 400-500 other web sites, this can REALLY smoke your results. Imagine GoogleBot crawling for you, finding that your server can’t/won’t respond quickly because your site is queued to respond behind a dozen or two of your co-hosted sites which have received requests for information at the same time as yours—maybe even from Google—and therefore sees you as being “slow”.

You’re now going to be rank-penalized because:

• A) you have a cheap plan (MAYBE fair) and
• B) at the moment Google came knockin’ your neighbors were creating traffic and making noise and so Google thought you weren’t home (NOT fair).

There’s some good news here: slow-performing web site elements will be cut back. The Flash-heavy aspects I referred to above are almost certain to become less and less prevalent, and sites that call dozens of advertising elements every time you load a page are going to have to change the way they do things or face the ire of Google as their SEO rankings drop lower and lower over time.

But if concern for how search engines see sites drives business process you might also find a bias toward going back to the way web sites looked circa 1999.

In the statistics Google is making available to me as of this writing, the three web sites I control have very different speed results.

• answerguy.com takes 1.6 seconds to load a page (82nd Percentile)
• pc-vip.com takes 0.3 seconds (99th Percentile)
• hatetexting.com loads a page in 0.9 seconds (95th Percentile)

Why the disparity? Here are a few points:

1. Answer Guy Central and PC-VIP are hosted on the same machine, while I Hate Texting is hosted elsewhere, so we know it isn’t just the machine creating the results.
2. PC-VIP is the only one of the three web sites  that uses (minimal) Flash, Answer Guy uses a Content Management System, and I Hate Texting is as simple as simple can be. This says that I Hate Texting ought to be the fastest, but it’s right in the middle.
3. PC-VIP shouldn’t manage to be as fast as it is, but because the measurement is “per page” and what looks to a human like one page is actually three in the eyes of a search engine delivering a measurement, its numbers are amazing.
4. Answer Guy, fully 500% slower than PC-VIP, still manages to load a tremendous amount of information at a speed that’s faster than 82% of all web sites.

Given how much information is loaded when you ask for a page here at Answer Guy Central, I’m pleased that it takes “only” 1.6 seconds to load. But what if Google decides that’s “slow”?

Is your head spinning?

Keep an eye on this; it could have a big impact on the way you’re seen on the Internet. Contact Us if you want help. And please: when you work on your SEO plans, make sure you take everything into account.

In a nutshell, the idea is that you don’t really have competitors any more. Now, coopetition has created a business environment where you find a way to work with others who formerly looked like competitors, but are now viewed as collaborators—even if there’s a zero-sum game for your shared potential customers.

Coopetition isn’t all that new an idea, but it’s the Internet and our new incredibly short attention spans that’s finally brought it to an easily explained place. I mentioned last week that CNet Managing Editor Jon Skillings had actually asked me to read their material and take the extra time needed to comment, but not to identify myself. Ridiculous. In adding (invited!) opinion to a post you add value to it both by expanding the readers’ minds and by keeping the readers on the post’s web site longer.

So here’s the next step: there’s a piece of code that can be added to Internet content that tells Google and other search engines that they should ignore links. That code is called NoFollow, and I can’t think of a good reason to use it.

OK, I can, but it doesn’t work when business change and coopetition are brought into play. NoFollow doesn’t stop a link from working, so if you comment on something you find on the Internet and include a link to back up your opinion that link will still do what you expect. But it does enable the site adding NoFollow to links to make your opinion “not count”.

And of course the reason to do that is to maintain a position of superior influence. In cases where you’re trying to cut down on SPAM, NoFollow could have a place, but there are other tools to handle that and I promise you every big web site uses them. And they work better than NoFollow

I’m all for winning. But NoFollow isn’t creating a long-term win, and as people come to understand that and call its users on the fact that they are saying “we want your opinion here to make us look important, but it doesn’t count” it’s going to backfire.

NoFollow: Don’t use it. And don’t stand still for other who short-circuit business change by doing so.

OK, so the truth here is that ten out of fifteen people in a study commissioned by Microsoft thought so. The data may be dirty for any number of reasons, and the sample is too small to take seriously. But the idea is intriguing. If Microsoft convinces enough people to try out Bing, will a huge percentage of them abandon the Google ship?

Business Change at its not finest.