Stepping Back From the Flash Indexing Hype

Julie Batten recently wrote a prescient column that described a healthy perspective on Flash use in enterprise Web sites. In the column, Julie described a client’s indecision about heavy Flash use based on conflicting advice.

Little did Julie (or any of us) know that within a few weeks, Adobe, Google, and Yahoo would team up to make Flash searchable for crawling and indexing. While reading the announcement, I pictured Julie’s client trying to reach her frantically on the phone, more confused than ever due to the new information he had to weigh.

Existing Examples of Flash Indexing

Despite last week’s announcement, indexing Flash files isn’t exactly new. Google has returned Flash files in results for quite a while, but the results haven’t exactly been spectacular.

For example, Hewlett-Packard has an excellent timeline that details the company’s history, including information about the founders, products, and earnings. When I searched for an exact quote from the Flash movie, in this case “oscillators for Walt Disney,” this Flash file was the third result. This isn’t the URL of the timeline page. It’s the URL of the timeline itself, which is a Flash movie embedded into the timeline page.

Clicking on the Flash file’s URL, you can see some of the dangers inherent in extracting Flash content from within HTML wrappers:

  1. The Flash movie has expanded to fill the browser screen, which makes the images and text appear larger and a bit jaggy. Not a huge distraction in this case, but not what the developers had in mind.
  2. There’s a total lack of navigation outside the timeline itself without the HTML wrapper code. True, you can move from the default 1930s era to the 1990s as intended, but there are no breadcrumb or typical navigation links that let you maneuver into different parts of the site.
  3. While the request for the Flash file will likely show up in HP’s server logs, common JavaScript-based analytics programs likely won’t show the referral information correctly. I could come from Google, spend an hour looking at the Flash file, and if that file has no analytics triggers, any subsequent page I go to won’t show that I came from Google. Of course, it’s unlikely that I’ll go to any other HP pages during the same session anyway, due to the navigation issues.

To be fair, I used a quoted text string as query text, which relatively few people do. So that limits the number of users who will see a result like this, at least for now. But I think it’s fair to expect that as crawling of Flash content continues, typical, non-quoted query strings will begin to show Flash content in the SERPs (define) as well.

To React or Not to React

Last week’s announcement about Flash resulted in a record number of clients writing me to ask if I’d read about it, which means this story had greater than usual penetration into traditional marketing publications. It’s big news, especially if you take the headlines at face value.

I told my clients to hold steady and not radically change course anytime soon. We’ll see how the SERPs shake out, and closely monitor the evolution of Flash content showing up in results. We’re not quite ready for primetime yet. Within the next few years, however, I’m hopeful that developers and engines can agree on ways to present the content that both creates a useful user experience and addresses the concerns I’ve laid out here.


Headlines make the SEM (define) world go round, and they maintain the buzz supply for those who feed and thrive on it. But pragmatism demands a more cautious approach. Continue to perform random searches in your vertical and look at what the engines return. Base your architecture decisions on that, not what you read in the tech headlines.

Related reading

Brand Top Level Domains