I've Watched Traffic Die Three Times

I have now watched search traffic die three separate times on sites I am responsible for. Not dip. Not wobble. Die, in the way where you stop refreshing the chart because you already know. Each death had a different cause, and each one taught me something the last one didn’t.

I want to write it down while I still remember how each one felt, because the people selling recovery courses tend to smooth all three into one story. They were not one story.

Wave one: the Helpful Content Update

The first was Google’s Helpful Content work, landing hardest through late 2023. This one felt like a judgement. Pages that had ranked for years were suddenly told they were not helpful enough, and the traffic left over a few weeks. On one site I manage, a big chunk of steady organic traffic went and did not come back at the old level.

I was not the only one. Independent trackers documented losses across hit sites commonly in the 40 to 80 percent range, and most affected sites had still not recovered many months later. One analysis of 670 travel sites found four in five lost traffic, nearly a third of them by 90 percent.

What it taught me: a lot of what the industry called “good SEO content” was thin, and the update was not wrong about all of it. Painful, but not entirely unfair.

Wave two: AI Overviews

The second was AI Overviews, from mid-2024. This one did not feel like judgement. It felt like the floor moving. The rankings were often still there; the clicks just stopped, because the answer now sat above them on the page. You could be number one and watch the visits fall anyway.

The data backs up the feeling. Ahrefs studied 300,000 keywords and found the top-ranking page loses about 34.5 percent of its clicks when an AI Overview sits above it, and their later data put the drop closer to 58 percent. Pew Research, watching what real people actually do, found searchers click a result on just 8 percent of searches that show an AI summary, against 15 percent without one. Roughly half the clicks, gone.

What it taught me: ranking and traffic quietly became two different things. You can win the position and lose the visit in the same quarter.

Wave three: the chatbots

The third is still happening: people asking an assistant instead of opening a search engine at all. This one is the hardest to measure, because the visit that never happens leaves no trace in your analytics. You feel it as a slow leak in the top of the funnel that no report fully explains.

The scale is still a matter of forecasts, but they are not gentle ones. Gartner expects traditional search volume to fall 25 percent by 2026 as chatbots and virtual agents absorb the questions people used to type into a search box.

What it taught me: the scariest decline is the one your tools can’t show you.

Why I keep managing sites anyway

Because all three waves killed a certain kind of traffic, not a certain kind of business. The sites that had a real reason to exist, a genuine product, a reputation, an audience that wanted them specifically, took the hits and kept going. The sites that were only ever a pile of ranking pages did not.

That is the actual lesson across all three deaths, and it is almost boring: be a real thing. I have watched the alternative fail three times now.