Recent content by MadOnGrind

The keyword check a few posts up is the bit most people skip, agreed. One thing I'd add on top of it: check the actual page the link will sit on, not just the domain. I've had links on sites with genuinely good niche traffic where the specific post was four clicks deep off the homepage, nothing...

The thing that caught me out on a cleanup like this was assuming low traffic meant low value. I pruned a batch of flat pages on an older site once and two of them turned out to be the only URLs anything external was pointing at. Undoing that took longer than the cleanup did. So before removing...

The bit that catches me out every time is not the mapping, it's what happens after. I can draw a lovely pillar and twelve supporting pages, and three months later half the supporting pages are getting crawled once a quarter because the only links pointing at them sit in one paragraph on the...

The "brand new + 20k pages" combo is probably the whole story. Crawling 20k URLs costs Google something, and on a domain with no history it has no reason to spend that yet. One thing nobody's mentioned: tube setups usually spin up huge numbers of near-identical URLs on their own (tag pages...

On the "stick with long articles because AI answers the quick questions" point, I'd push back a little. The times I padded a page out to cover everything, the thing that actually moved was splitting it instead. Had a page trying to answer a "what is" question and a "how do I pick one" question...

The thing I'd chase first is the 1,400 server errors, not the quality question. Everyone's telling you it's a trust thing, and maybe it is, but "trust" isn't something you can go and look at. 5xx is. Pull raw server logs for Googlebot only, a couple of weeks of them, and group by status code...

The point earlier about splitting the drop by query is where I'd start too, but I'd add one layer under it. In GSC, check whether the page lost position or lost clicks at the same position. Those look identical in a traffic graph and they are not the same problem at all. If average position...

On the title tag point above, that one is underrated. On my end the version that worked was pulling the pages in GSC with impressions but almost no clicks, then rewriting only those titles to match how people were actually phrasing the query. No new content, no links. A couple moved within a...

Relation-based grouping like that works well for local service sites. One thing I'd check in your setup: whether the insects tab itself sits on a route crawlers actually travel, or only in a dropdown nav. I had a client where the whole silo hung off a menu item and log files showed it barely got...

@Bigslap the clicks part makes sense to me, because short pays twice there. The tight answer paragraph that wins you the click is the same one a longer guide would have opened with anyway, so the short version isn't really missing anything the searcher came for... it just skips the padding...

@MrCompEng before rewriting a single title, check where those impressions actually happen. 150 a month is about five people a day seeing your listing anywhere in the results. Even a perfect title turns that into a click or two a week, so the CTR advice in here is a bit early for where the site...

@SyedHabib-SEOExpert a staging noindex making it to production might beat my missing tag for silliest false alarm. Yours at least looked like a real problem from the outside. Funny how both failures point the same way though. The graph looks dramatic, so the head goes straight to penalties and...

A while back I opened analytics on one of my sites and traffic had fallen off a cliff. Overnight. I did what I suspect most of us do... assumed Google had finally come for me. I spent longer than I'd like to admit on ranking theories. Re-read the update chatter, started listing pages to audit...

@ClivoraX one source nobody here has mentioned yet... your own site search. The queries people run once they're already on your site are a gap list your actual audience wrote for you, and the ones that return no results are the posts you should have written already.

@astridmonroe that trap is the right way to put it. Writing to fill a slot rather than because there is something to say. What took the edge off for me was letting an update count as a publish. A post already picking up impressions usually needs an hour rather than a day, and it tends to move...