On Monday, Google announced that “we’ve greatly improved our ability to index Flash.” Since then I’ve heard and seen a lot of excitement flying around the web via articles, comments, instant messages, and even in my Facebook. At the onset, this “upgrade” sounds great, but let the buyer beware. You may not be getting what you think you are. There are potential positives to come out of Google improving their ability to index Flash. However, there are many unanswered questions that scare me any many of us in the SEO community.
This morning my coworkers and I talked about this in great length. Does this effect how we design and build websites? In the end we agreed, without any debate, that it does not. Here’s why:
“Improved” is a Scary Word
“Improved” does not mean “fixed” …and the distance between the two can be inches or miles. There is still a lot of confusion around what this improvement is. Does it mean they can simply read more text? Probably. Does it mean they can follow all links within a flash file? Perhaps. Does it mean they see each “page” within the Flash file as a separate indexable page? Not likely.
The Entire Flash Site is Still, Most Likely, Seen as One Single Page
Even Google has the ability to read all content and links within the file, there is a very good change that they see it as connect to one single URL. I think this comment from mike.tekula at SEObook.com says it best:
While I see this as a positive step in some regards, I also see it potentially causing a lot of confusing for people who don’t fully understand search. Googlebot indexes pages, right? Pages have unique URLs, right, and unique title tags? How many full-Flash websites do you see where the URL changes as you navigate from section to section? How about the title tag?
So what is Google going to index?
Building a site completely in Flash, whereby only one file loads and the URL/Title doesn’t change as the user navigates, is still going to be a shot in the foot for anybody hoping to pick up a lot of search engine traffic. Not to mention the usability issues…
To boil it down: If, let’s say, a 100 “page” site is built in Flash and sits on one URL …what is Google going to index? How likely is it going to be that Google will display a result at the top of their of their search results that sends visitors to a Flash homepage that does not appear, to the user, to be related to what was searched? My guess is, not very likely.
What About the other 5 Billion U.S. Searches per Month?
Although Google makes up about 65% of all U.S. searches, it’s just bad business to only build just for Google. There’s still Yahoo, MSN, Ask, & AOL who, although are not the big dogs, still account for 1.5 million searches per day in the U.S. alone. Starbucks is still going to sell tea, water bottles, & snacks, even though 95% of the customers just want some kind of coffee drink. We shouldn’t discount the other engines.
Accessibility
One of my co-workers, Patrick, brought this up …and it’s a very good point because good accessibly = good SEO. The more accessible a page is to a user, the more it is accessible to a search engine …and just like my Starbucks example, you don’t want to limit your audience.
My position has been, and still is, that Flash should only be used where you would normally use an image or a video …not as a collection of “Pages” within a single URL. In other words:
- This is the right way: Disney Cruise Line
- This is the wrong way: Disney Parks
Credits: Google/Flash image borrowed from the creative genius, Roy Tanck
Categories: Search Engine Optimization, Search Engine Marketing ~ ~ Trackback

July 2nd, 2008 at 1:51 pm
You’re right, improved does not mean fixed! Stay away from flash sites period! Nothing but SEO hell…
July 2nd, 2008 at 7:14 pm
Good post. As I understand it, the new player *does* let Google follow links within a Flash file. And it can detect what’s happening to the Flash movie. So in theory, if you’re using a Flash deep linking technique (like SWFAddress or the Deep Linking in the Flex Framework) Google might be able to detect that the URL has changed and would be able to associate the content with that specific URL and then index that part of the application.
We’ll see if it actually works that way.
=Ryan
rstewart@adobe.com