How much of a ranking drop do you actually wait out before assuming it's an algorithm update vs something you (or a client) broke?

Every time rankings dip, there's this internal debate: is this a core update, a technical issue, seasonality, or something someone touched on the site without telling me? Some people say give it 2 weeks before panicking; others start diagnosing day one.

What's your actual rule of thumb before you start pulling apart the site looking for a cause?

I think it's better to look at the bigger trend instead of reacting to every small ranking change

Wait about a week unless the drop is really significant. Small ranking fluctuations are normal, so I check for technical issues first and then see if it's affecting traffic before making any big changes.

One bad day doesn't always mean something is broken I'd wait for a clearer trend

still new to this but what helped me was realizing it's not really about how long you wait, it's about checking if it's just you or everyone, i look at semrush sensor/seroundtable first, if the whole SERP is shaking that day it's probably an update and i just leave it alone for a week or two. if the trackers are calm and only my site dropped, then i probably broke something, so straight to search console checking indexing and manual actions

I usually do not judge it by the size of the ranking drop alone. The first thing I check is whether there were any recent changes to the website, such as content updates, redirects, technical changes, or new robots settings. If nothing has changed and I see similar movement across multiple websites, I am more likely to suspect an algorithm update.

SK_LinkBuilder said:

Some people say give it 2 weeks before panicking; others start diagnosing day one.

Click to expand...

The duration question is the one I would skip. Shape and spread tell you more than how many days you wait.

Pull the last 90 days in Search Console and compare page by page and query by query instead of looking at the site total. A sharp fall on one date that hits everything is usually external. A fall confined to one template or one class of query is nearly always something that changed on those pages, and the site total hides exactly that.

Worth checking the change history yourself before asking whether anything changed, too. Nothing changed is the most common answer and it is rarely accurate.