Multimedia Metadata Optimization

Image and multimedia SEO (define) is slowly evolving. Just look at Google’s, Ask’s, and Yahoo’s, main pages. All have image search tabs, and Google and Yahoo have video tabs as well. As a visual person (I’m a designer/developer as well as an SEO expert), I always get excited when I see search engines evolve beyond text-based information retrieval. Plus, I love Flash. Give me a good usability reason to use Flash and, baby, I’m there.

After my multimedia buzz wears off, the logic part of my brain kicks in. Since image and multimedia search are still in their infancy, how should SEO professionals optimize these file types? Let’s start at the beginning. SEO professionals and Web site owners must consider file-type optimization as part of the SEO process.

What Is a Document?

Many people have preconceived notions about the definition of “document.” Heck, even I did, especially when I thought of a Web document. Most people think of a document as a piece of paper, or a file, that contains text. Currently, information retrieval systems are primarily text-based. In other words, a commercial Web search engine crawls the Web and retrieves text-based information, which is the reason using keyword-rich text on a Web page is very important.

Documents aren’t necessarily text-based, however. In a museum, the actual object being displayed, such as a mummy, could be considered a document, for example. When you type “King Tut” as a Google query, Google doesn’t present King Tut’s mummy in search results. Instead, Google presents a text-based substitute. This text-based substitute is a document surrogate, because Google can’t currently retrieve mummies and display them as search results. (Wouldn’t it be really cool if it could?)

SEO professionals have been working with document surrogates for a long time. Meta tags are one example. In the event a search engine can’t retrieve information about a Web page, such as a graphics-intensive Web page, it often uses the information about the page from the meta-tag description and descriptions of the page from text-based Web pages.

Multimedia SEO is all about optimizing document surrogates, not the actual document. As multimedia search evolves, search engines will be able to determine a multimedia file’s actual content. Until this evolution is complete, search engines rely heavily on metadata. That’s where SEO professionals, librarians, and other information specialists become key players in effective information retrieval.

Meta Tags or Meta Garbage?

Back in the late ’90s, few people knew what meta-tag description and keyword attributes were or how to effectively use them on Web pages. These attributes were once known as SEO’s secret weapons because many Web pages would rank well on popular commercial search engines at the time (Inktomi, AltaVista, Infoseek, etc.) just by keyword-stuffing HTML title tags, meta-tag description attributes, and meta-tag keyword attributes.

A keyword-stuffed meta-tag description might look like this:

A keyword-stuffed meta-tag keyword list might look like this:

The problem with these metadata examples is keyword stuffing, meaning unnecessary over-repetition of query words. Because Web-page metadata had a commercial effect (high search engine positions), they’ve been exploited, the main reason meta tags have little or no effect on text-based Web files in SEO now.

Yet there’s still a huge need for metadata for non-text Web files, and meta information for audio, video, and graphic image files is still used to determine relevancy on the multimedia search engines.

How much metadata is genuine information about an audio or video clip and how much is garbage, especially since SEO professionals want their client sites to rank well in multimedia search engines as well as the text-based search engines?

Meta-Tag Lies

Many SEO professionals lie about meta-tag content. They also put invisible text on invisible layers so search engines see content users can’t. Many create thousands upon thousands of fake Web sites to mimic link popularity. The current craze is to buy links on publisher and news sites, links that aren’t always easily distinguishable as paid advertising links. Personally, I’m patiently waiting for a major search engine to ban a news or publisher site for this practice. It will happen.

Yes, poor Web-page metadata provided immediate commercial benefits in the past. It also meant site developers didn’t have to create user-friendly Web sites to obtain search engine visibility. The result, though, was commercial search engine users were presented with low-quality, non-user-friendly sites at the top of search results pages.

Fortunately, there’s a movement toward more user-centered site design. However, few usability professionals understand the search arena. And few search experts understand site usability.

In the meantime, I’m watching the multimedia search arena slowly become exploited, just as I watched meta-tag descriptions and keywords exploited back in the late ’90s. I have to admit I’m confused. Why are the commercial Web search engines allowing this to happen? Didn’t they learn from the meta-tag experience in the late ’90s?

Conclusion

I love the entire SEO/SEM (define) business. Our group has truly revolutionized information retrieval, especially in getting people to understand the entire keyword research process as well as using audience-based keyword phrases on Web pages.

But I don’t understand how search engines can remain so naive about metadata and multimedia search. I certainly didn’t write this week’s column to encourage meta-tag spam. Regular readers certainly know how I feel about that.

I wrote this as a wakeup call to the search engines. Quit making the same mistakes. Get a better handle on metadata optimization. It will make for a better search experience for all of us.

Want more search information? ClickZ SEM Archives contain all our search columns, organized by topic.