The Rise of Keyword Research
How keyword research became its own discipline, from WordTracker's 1998 launch to search engines abandoning the meta keywords tag.
A few years ago, “keyword research” wasn’t a phrase anyone used. Deciding what words to put on a page was mostly guesswork. That’s changed quickly, and it’s worth tracing how.
The meta keywords tag era, and its collapse
In the web’s early years, the HTML meta keywords tag let a page author simply list the terms they wanted to rank for, and early engines took that list more or less at face value. It didn’t take long for site owners to abuse it, stuffing tags with hundreds of irrelevant terms to capture unrelated traffic. By the late 1990s, engines including Infoseek and, increasingly, Google are discounting or ignoring the tag entirely, forcing a shift toward analyzing actual page content and link structure instead of self-reported labels.
WordTracker gives keyword research a data source
WordTracker launched in 1998 as one of the first tools built specifically to show real search query data, pulled from metacrawler engines like Dogpile that aggregate results from several smaller engines at once. For the first time, a site owner could see, roughly, how often a given phrase was searched, instead of relying purely on intuition about what customers might type.
GoTo.com’s bids become an accidental research tool
GoTo.com’s pay-per-click model, launched in 1998, requires advertisers to publicly bid real money on specific keywords, and those bid amounts are visible to anyone browsing the results. That transparency turns GoTo.com into an unintentional keyword research tool in its own right: a term with many advertisers bidding aggressively is a term with proven commercial demand, something no query-volume tool alone can tell you.
Forums turn scattered tricks into a shared practice
Communities built around search marketing, including Search Engine Watch, founded by Danny Sullivan in 1997 to track how the major engines work, start giving practitioners a shared vocabulary and a place to compare notes. What used to be trial-and-error tricks passed around informally are becoming a documented, discussed practice, with keyword selection treated as a distinct skill rather than an afterthought tacked onto page design.
Where the discipline is heading
Keyword research in 2001 is still closer to a craft than a science, built on a patchwork of query-volume tools, bid transparency and page-level testing rather than any single authoritative source. But the direction is clear: businesses that treat word choice on a page as a deliberate, researched decision are outperforming those that write copy purely for internal taste. That shift, from guessing to researching, is the real story behind keyword research’s rise, and it’s one that’s likely to keep deepening as more tools and more data become available.
More from 2001: browse the SEO archive or see everything published this year.
