Google emailed me about my website. Here's how to tell whether anything is actually wrong.

It arrives sounding like a warning from your bank. In your inbox it shows up as this, with your own address at the end:

The subject line

New reasons prevent pages from being indexed on site …

Open it and the heading inside says something slightly different — worth knowing, because these are the words most people end up searching for:

The heading inside the message

New reason preventing your pages from being indexed

Same message, two spellings. It is automatic, it also waits for you in Search Console behind the bell icon at the top, and inside it names one or more reasons — short phrases like “Page with redirect” or “Not found (404)”. If you have never seen one before, the natural reading is that something broke and it is your fault.

Usually neither is true.

What Google is telling you

Google keeps a list of your pages — the Page indexing report — and, for each page it decided not to put in search results, a reason. The message says so in its own words:

The message body

Search Console has identified that some pages on your site are not being indexed due to the following new reasons:

Note that the heading says reason, singular, while the sentence beneath it adapts — one site of mine got a message naming two at once. Either way, it fires when a reason appears on that list that wasn’t there before — not when things get worse, and not when a lot of pages are affected. Just: a new category showed up.

That distinction matters more than anything else in the message, because it means:

• One page can trigger it. The message reads identically whether one page or four thousand are affected. It names the reasons and nothing else — no counts, no addresses.
• It can fire for something you did on purpose. Marked a thank-you page as “don’t list this”? That’s a new reason, and Google will write to you about it. Google knows this, which is why its own advice in the message is conditional: “If any of these reasons are not intentional, we recommend that you fix them.”
• It is not a penalty. Nothing has been taken away from you. Nobody at Google looked at your site and disapproved.

The reasons are the whole message

Everything useful is in those short phrases. Here is what each of the common ones means, and whether it is asking anything of you.

• Nothing to do “Page with redirect” — an old address forwards to a new one. That is what redirects are for. If you moved to https or dropped the www., this is expected.
• Nothing to do “Alternate page with proper canonical tag” — Google’s own message contains the word proper. It found a duplicate and you had already told it which version is the original. This is Google saying you got it right.
• Nothing to do “Discovered — currently not indexed” — Google knows the page exists but hasn’t looked at it yet. Usually a matter of waiting.
• Worth an hour “Excluded by ‘noindex’ tag” — a page carries an instruction telling Google not to list it. If that page is one you want found, this is your answer, and the fix is usually one setting in whatever you use to edit your site — and that setting has its own article, because it is rarely one you remember touching. (Google’s help page calls this same reason “URL marked ‘noindex’” — the interface and its own documentation disagree on the name, which is worth knowing before you go searching.)
• Worth an hour “Blocked by robots.txt” — a file on your site is telling search engines to stay out of those pages. Sometimes deliberate, often left over from when the site was being built.
• Today “Server error (5xx)” — your site failed to answer when Google asked. If this one has a real number next to it, tell whoever hosts your site.

There is one more that belongs in neither column. “Crawled — currently not indexed” means Google looked at the page and chose not to add it. No error, no rule broken, no instruction to follow — it just didn’t think the page was worth listing. That one has its own article coming.

So do you need to do something?

Two questions, in this order.

Which pages? Press the message’s Open indexing report button, or go to Indexing → Pages in Search Console and find the row matching the reason. Click it. You get the list of affected addresses.

Now look at that list honestly. Are these pages you care about — your home page, your services, the things you’d send a customer to? Or are they tag archives, filtered listings, an old thank-you page, a duplicate print version? A large number next to a reason means very little. Twelve throwaway URLs is nothing. One home page is everything.

Is the reason one that needs a person? Use the list above. If the answer is “pages I don’t care about, for a reason that isn’t a fault”, then you are done. Close the message. That is a real answer, not a way of avoiding one.

The part where it gets slippery

Say you did find something and you fixed it. You unticked the setting, or corrected the page.

Search Console has a Validate fix button that asks Google to check. It’s the obvious next move, and it’s worth pressing. But read how Google’s help page for that report defines the successful outcome:

How Google defines a passed validation

All known instances of the issue are gone (or the affected URL is no longer available).

Read the bracket again. A page that stopped existing counts as fixed. Delete the page instead of correcting it, and validation passes. That isn’t Google being careless — it’s an accurate description of what a report can observe from the outside. But it means “the warning went away” and “the problem is solved” are two different statements, and the report can only ever confirm the first one.

Meanwhile validation itself takes its time. Google’s own estimate is up to about two weeks, and sometimes much longer. It is easy to never press the button at all, because nothing in the interface explains what it does. And if it comes back “Failed”, that has its own article — the short version is that one page is enough to fail the whole batch.

How to actually confirm it

There is one free check that works right now, and you should use it before anything else. In Search Console, paste the page’s address into URL Inspection at the top and press Test live URL. That fetches the page as it exists this second and tells you whether indexing is allowed on it. If it still says blocked, your change didn’t land — and you learned that today instead of in two weeks.

What it can’t do is compare. It has no memory of what the page said last week, so it can’t tell you whether anything changed, only what is true now. And it answers about one page at a time.

That gap is the one worth knowing about. A live test tells you what is true on one page right now. It cannot tell you whether that is different from last week, and it cannot answer for the site as a whole.

Written by Oleh Koval at Specia1ne, who reads these reports for a living.