I told a client to keep an eye on the next Google update. Said it almost offhand, in a monthly call, because the traffic trend had finally turned and that's usually when updates get interesting. The next day, Google rolled out the August 2026 Spam Update.

I wish I could tell you I planned the timing. I didn't. What I'd actually done was spend the last 6 months getting the site into a state where a Google update had something good to boost.

This is a large e-commerce site, and the surface-level story was a slow, steady traffic decline that nobody could pin on one obvious cause. They came to us after a pretty fresh domain migration.

The drop was compounding.

Crawling was inefficient, with the bots spending their budget on pages nobody needed.

Indexation was messy, with thin and outdated pages competing against the ones that actually mattered. And rendering issues were also causing Google to crawl empty pages.

Those issues combined made the site's actual best content fight for attention against a huge amount of dead weight.

The real weight on this site was out-of-stock products that had been left live for years. Over 100,000 pages.

Those products were still bringing around 20-30% of the traffic. But when I dove into this traffic, I found that it was worthless - sessions lasted 2-3 seconds. No purchase intent behind them anymore. They generated around 0.3% of the revenue of the site, and they were covering around 85% of the entire structure. They were sending bad-quality signals across the whole domain.

The obvious fix sounded insane but we did it. We redirected roughly 80% of the site.

I understand why that number makes people nervous. It made the client nervous. Cutting four out of every five URLs on a live e-commerce site feels like the kind of decision that should require six approvals and a backup plan for your backup plan.

But the pages we were cutting weren't generating revenue, weren't generating rankings. They just diluted quality signal sitting on top of the pages that actually mattered.

(In the meantime, we also fixed rendering, site structure a bit and crawling)

After this change, the traffic dropped - yes, we lost another 30% of traffic when the client was already losing traffic, and that’s exactly what we expected. We unfortunately did the change right before the Google update, so it also compounded on the drop.

The traffic went down, but revenue stayed the same. We lost only the bad quality visits and were left only with the SEO traffic that was actually converting - at least better. The SEO purchase rate doubled - the SEO user data started becoming more and more positive, and we wanted to see how it impacts the traffic.

After a couple of weeks, we saw the trend reversing - week after week, our site was gathering more and more traffic.

We started recovering what we lost, positive user data started reversing the traffic trend, and we finally stopped the negative trend that was killing this site. That’s when I decided to drop the bomb.

I saw it time and time again that such long-term positive trends make the next updates interesting - I sense it in the air and I felt it here too.

I talked to the client on the next call that we can expect an interesting jump in traffic during the next Google Update.

I was not expecting to be right about the timing. The August 2026 Spam Update rolled out the day after that conversation, and over the following days the site's traffic grew by 30% to 50%.

I want to be precise about what I think actually happened, because it's easy to tell this story as "we called the update" and that's not quite it.

What we did was remove the majority of the negative signals dragging the domain down, months before the update ever shipped. The update rewarded work that had already been done and had already started compounding on its own.

A site heading into an update bloated with dead stock and crawl waste is walking into it exposed. A site that's already cut the fat and cleaned up its rendering is walking into it with something worth boosting.

This is also, not coincidentally, the whole argument for why we don't sell static audits. An 80 percent site reduction isn't a recommendation you hand over in a PDF, and hope gets picked up by an internal team six months later. It's a mission you execute with the client's developers, verify week over week, and adjust as the data comes in.

The prediction I made wasn't a guess pulled from nowhere. It was a read on a trend we'd been tracking in real time, on a change we'd made together, on a timeline we controlled.

The part I keep coming back to isn't the 30 to 50 percent. It's the two months before the trend turned, when the only evidence I had that the work was right was the work itself.

Updates get all the attention because they're visible and dated and easy to point at. But by the time one lands, your site's fate in it was mostly already decided by what you'd done in the months before. The update is just the moment the decision gets read out loud.

Reply to this email with “Bloat” and your e-commerce domain.

I am selecting ONE site to do a free technical SEO analysis, including crawling and indexation, to see if your site is full of low-quality pages that kill your rankings.