In 2008 I Was Asked What the Next Big Thing in Search Was Going to Be.
Over my career I have occasionally stepped away from independent work to spend time inside large corporations — not because I needed to, but because the experience of operating at that scale teaches things you cannot learn any other way. In 2008 I was interviewing for a web marketing manager position at a large global company. Two rounds — the hiring manager in the first, his boss in the second. Both asked the same question: what is the next big thing coming for search engine rankings?
It was a smart question. The company was heavily dependent on competitive search terms and spending roughly two million dollars a year in PPC to stay visible. They needed someone who not only understood where search was, but where it was going. I held that position for three years and managed that budget — but the answer I gave in that interview is the one I want to talk about here, because it has not changed.
My answer was not a prediction. It was a history lesson. And the history lesson was this: text has been king since the first search engine that actually worked, it was still king in 2008, and it was going to stay king. I was right then. I am still right now.
Before Google There Was Excite. And Excite Changed Everything by Reading the Page.
I started gaming search engines the right way before Google existed. In 1995 the search engine that mattered was Excite — and what made Excite revolutionary was not the interface or the brand. It was the fact that it largely ignored meta tags and indexed the actual text on the page.
That distinction matters more than it might sound. Before Excite, search engines relied heavily on meta tags — the hidden descriptive fields in a page’s code that a site owner could fill with whatever words they wanted, whether those words had anything to do with the actual content or not. That system was being abused badly. Sites with no legitimate relevance to a search query were stuffing meta tags with popular keywords to capture traffic they had no business receiving. The signal was broken because it was too easy to fake.
Excite’s approach was different. It read the words on the page — the actual visible content — and used that to determine relevance. Suddenly the text you put on your site was not decoration. It was the ranking signal. Gaming the system the right way meant writing genuinely relevant, well-structured content. The sites that did that ranked. The sites that were stuffing hidden fields with irrelevant keywords stopped ranking.
That was 1995. The underlying principle has not changed in thirty years. Every major evolution in search since then has been Google and its successors finding better and more sophisticated ways to read, understand, and evaluate text.
Google Did Not Invent Text as a Ranking Signal. It Just Got Better at Reading It.
When Google arrived it built on the same foundation — text matters — and added layers of sophistication that made the signal harder to game and more meaningful as a measure of genuine relevance. PageRank added the dimension of authority: not just what the text said, but how many other pages with relevant text pointed to it. Later updates added context: not just whether a keyword appeared on the page, but whether the surrounding content demonstrated real understanding of the topic.
Each of those evolutions was Google getting better at reading text. The algorithm updates that sent shockwaves through the SEO industry — Panda, Penguin, Hummingbird, BERT, the Helpful Content updates — were all, at their core, Google refining its ability to distinguish between text that genuinely informed a reader and text that was written to manipulate a ranking. The punishment was always for the latter. The reward was always for the former.
The professionals who panicked at each update were the ones whose strategies depended on exploiting weaknesses in how Google read text. The ones who were unaffected — or who benefited — were the ones who had been writing genuinely useful, well-structured content all along. The game kept changing. The winning move never did.
- Panda — penalized thin, low-quality, and duplicate text content
- Penguin — penalized manipulative link schemes built around keyword-stuffed anchor text
- Hummingbird — improved understanding of natural language and conversational text queries
- BERT — enabled Google to understand context and nuance in text, not just keywords
- Helpful Content — rewarded text written for people, penalized text written for algorithms
Even Video Is Text. Google Reads the Transcript.
The most common challenge to the text-is-king argument is video. Visual content, people argue, is the future of the web — and if Google is ranking video results prominently, surely the era of text dominance is ending. It is a reasonable observation that leads to the wrong conclusion.
Google owns YouTube. And YouTube — along with Google’s broader video indexing capabilities — automatically transcribes video content and converts spoken word to text. What Google is indexing when it evaluates a video for search relevance is, in large part, the transcript. The words spoken in the video become text. That text is what gets evaluated, ranked, and surfaced in search results.
Video is a powerful format and it absolutely belongs in a well-rounded content strategy. It increases engagement, extends time on page, communicates things that text alone cannot, and reaches audiences who prefer to watch rather than read. None of that makes it a replacement for text. It makes it a complement to text — and the mechanism by which Google understands and ranks it is, underneath everything, text.
There is no format in the current search landscape that escapes the fundamental truth: Google is a text-reading machine. Everything it indexes, evaluates, and ranks passes through that lens first.
Text Is Not Just Content. It Is Structure, Hierarchy, and Signal.
Saying text is king does not mean writing more words and hoping Google notices. The quality of the text, the structure it sits in, the hierarchy it communicates, and the intent it serves all matter as much as the words themselves. A page with ten thousand words of loosely related content will not outrank a well-structured page of one thousand words that clearly and completely answers the question a searcher is asking.
Headings communicate hierarchy — they tell Google which ideas are primary and which are subordinate. Internal links with descriptive anchor text communicate relationships between topics. The presence of related terms and concepts around a primary keyword tells Google that the page understands the topic rather than just targeting a phrase. Page titles and meta descriptions are text signals that influence both ranking and click-through behavior.
All of it is text. All of it matters. And all of it has to work together in a structure that serves the reader first — because Google has spent thirty years getting better at recognizing when it does not.
- Heading structure communicates content hierarchy to search engines
- Surrounding context tells Google whether the page understands the topic or just targets a keyword
- Internal link anchor text signals the relationship between pages
- Page titles and meta descriptions are text that influences both ranking and click-through rate
- Quality and relevance outrank volume — one well-written page beats ten thin ones
1995. 2008. 2026. The Answer Has Always Been the Same.
I have been giving the same answer to the same question for thirty years. What is the most important thing for search engine rankings? Text. Well-written, genuinely useful, properly structured text that serves the person reading it. That answer has survived every algorithm update, every industry trend, every new format and platform and technology that was supposed to change everything.
The tools have changed. The sophistication of how Google reads text has changed enormously. The bar for what qualifies as genuinely useful has gone up considerably. But the fundamental signal — the text on the page — has never stopped being the primary currency of search visibility.
When I sat in that interview in 2008 and told a room full of executives that the next big thing in search was the same thing that had always been the big thing, I was not being evasive. I was telling them the most useful truth I knew: the businesses that would win at search over the next decade were the ones that committed to writing genuinely good content and building their sites around it. That turned out to be correct.
Text was king in 1995 when Excite started reading pages instead of meta tags. Text was king in 2008 when I staked a career move on that answer. Text is king in 2026. Invest accordingly.