The competition at the time was relying on HTML meta tags' keywords. It was purely based on faith that the web developer would be honest and help classify pages

The competition at the time was relying on HTML meta tags' keywords. It was purely based on faith that the web developer would be honest and help classify pages. Which was reasonable at the time, before the Web was so commercial, but that ship has long since sailed. Keyword-stuffing was already a thing that was happening, and other sites simply neglected to tag pages. Some search engines were starting to read the page content and count keyword frequency, so a page that had "apple" many times on it would be deemed relevant to a search for "apple." Google's whole thing was it did something smart, and treated the Web as the big graph it is and more or less measured the in and out degrees of the vertices to rank their popularity. The idea that something relevant would be linked frequently. This was the subject of an academic paper and had associated patents for awhile. https://en.wikipedia.org/wiki/PageRank https://en.wikipedia.org/wiki/PageRank It was a big deal at the time, but once again the Web has changed. Google's search relevance is a function of the changing Web and how it's now highly commercial and populated by everyone instead of being the world's largest community of nerds. Links aren't a measure of popularity, because gulf between the average user and people who operate web sites has widened...and the common denominator has dropped to the point that optimizing for popularity is questionable in the first place.