What holding #1 on Google for ten years taught me about SEO

For roughly ten years, my agency held the #1 spot on Google for "Chicago web design" and ranked Top 3 globally for "web design" — through Panda, through Penguin, through the mobile-first shift, through BERT, and now into an era of AI Overviews that has a lot of businesses convinced everything they know about SEO is obsolete. I watched competitors rise and fall across every one of those updates. The ones who fell weren't unlucky. They were playing a different game than I was, and that game only ever works until the next algorithm update makes it stop.

This isn't a nostalgia piece. It's the actual mechanism behind a track record that outlasted five distinct versions of how Google decides what deserves to rank — and why I'd bet on the same mechanism holding through whatever AI search becomes next.

Ten years, one spot, five different algorithms

Panda punished thin content. Penguin punished manipulative link-building. Mobile-first indexing punished sites that treated mobile as an afterthought. BERT rewarded content that actually answered the question behind a query, not just the keywords in it. And now AI Overviews reward content structured to be confidently cited. Each of these was, at the time, described as a seismic shift that would upend everything. Each time, the sites that had been built around gaming the previous version of the algorithm got hurt. The sites built around being genuinely useful barely noticed.

I didn't survive five algorithm eras by predicting each one. I survived them by never fully committing to a strategy that only worked because of a loophole.

What actually held the ranking

Strip away the specifics of any given year, and three things did the actual work, every single time:

• Genuine authority, not manufactured signals. Real depth on a topic, built over years, is expensive to fake and hard for any algorithm — human-built or AI — to mistake for something thinner wearing the same keywords.
• Consistency over bursts. A site that publishes steadily for years sends a completely different signal than one that publishes in a frantic push before an update and goes quiet after. Algorithms changed constantly. The reward for consistency never did.
• Usefulness as the goal, ranking as the byproduct. Every time I saw a competitor optimize for the ranking directly — keyword density, exact-match anchor text, content length for its own sake — they won for a quarter and lost the position within a year. Optimizing for the reader first turned out to be the more durable strategy, even though it never felt like the "SEO trick" everyone was searching for.

Every algorithm update in the last decade did the same thing: it got better at detecting the difference between genuinely useful and merely optimized. Betting on the second one was always a countdown.

What I abandoned along the way

To be clear, I didn't arrive at this discipline instantly — I ran plenty of the era-specific tactics that were standard practice at the time: keyword density targets, exact-match domain plays, directory submissions, and link exchanges that were considered best practice before Penguin made them a liability overnight. The pattern I noticed, watching those tactics get penalized one by one, is what eventually shaped the more durable approach. Every trick I abandoned had the same shape: it worked because it exploited a specific, temporary gap in how the algorithm measured quality. Once the measurement improved, the trick became a liability instead of an advantage.

Being an early Google partner during that period mattered less because of any special access, and more because it put me close enough to the actual guidance to notice how consistently it pointed away from tricks and toward fundamentals — long before "helpful content" became an official phrase in Google's own documentation. The businesses that treated that guidance as the whole strategy, rather than a hint to decode around, are the ones still standing now.

The mistake almost everyone still makes

The single most common error I still see, in every era including this one, is chasing whatever the current trick is instead of building the thing the trick is trying to approximate. Right now that mistake wears a new disguise: businesses flooding their sites with generic AI-generated volume, betting that more content equals more visibility. It's the same bet keyword-stuffing was twenty years ago, and it will resolve the same way — well, until the next update measures genuine authority even more precisely than the last one did, at which point the volume becomes the liability. I wrote more about exactly how that shift is already showing up in AI search results in SEO isn't dead. The businesses who think so are. — the mechanics of zero-click search and citation are the current chapter of the same story this post is telling.

The one thing I'd tell any business starting today

Don't ask "what ranks right now." Ask "what would still deserve to rank in five years, regardless of what the algorithm looks like by then." That single reframe would have kept most of the sites I watched get penalized over the last decade out of trouble, and it's the only piece of SEO advice I've never had to walk back.

The takeaway

A decade at #1 wasn't the result of finding the right trick — it was the result of refusing to depend on one. The tactics for expressing genuine authority changed completely, five separate times. The underlying principle — be worth ranking, not just structured to rank — never has, and there's no reason to expect the AI search era to be the first one where that stops being true.

Frequently asked questions