Short answer
On the rebuilt detmore.tr, Google indexed 65 pages three days after the report’s first record and 70 pages after a week. At the end of week three, 118 addresses were not indexed; 99 of them were there because of redirects and canonical tags, and the only real problem was a single 404. Three weeks brought 14 impressions.
On the morning of 6 November 2026, the Pages report in Search Console showed two numbers: 70 pages indexed and 118 addresses not indexed. For detmore.tr, which we rebuilt from scratch on the same domain in October, that looks like a poor school report at first glance; most of the addresses Google knows about are not in the index. Rather than write a general answer to how a new site gets into Google, we are opening our own records. These records cover what the report showed day by day in the first three weeks, what those 118 addresses were, where we found the one real problem, and what has not happened yet.
Every number here comes from detmore.tr’s Search Console property, from the export of the Page indexing report, and from the site’s own records. It is three weeks of one site; it should not be read as an average or a promise for any other.
The starting point: same domain, new addresses
The old detmore.tr kept every page under /tr/… or /en/…, and its addresses differed from today’s; for example, /tr/hizmetler/web-development now lives at /hizmetler/web-ve-yazilim/. On the new site the Turkish pages sit at the root and the English pages under /en/. The domain stayed the same, but for Google this was a site move with changed addresses. We did four things as part of the launch:
- Redirect map. Every address on the old site redirects with a 301 to its counterpart on the new one. Google’s site move documentation also recommends permanent, server-side redirects from the old addresses to the new.
- Sitemap. It held 72 addresses on day one. We submitted
sitemap-0.xmlin Search Console. - Search Console and Bing. Alongside Google, we verified Bing Webmaster Tools as well.
- IndexNow. We turned on Cloudflare’s Crawler Hints. When content changes, this feature notifies the search engines that support the IndexNow protocol.
How to prepare those steps is covered in keeping SEO through a site rebuild. Here we record what came after the launch: what the report said.
Week by week: what the report showed
The numbers the report showed from 17 October to 6 November fall into four periods.
| Date | Indexed | Not indexed | What happened |
|---|---|---|---|
| 17 to 18 October | no data | no data | The report is still empty |
| 19 October | 0 | 2 | Google recorded its first two addresses |
| 20 to 23 October | 65 | 58 | The first big jump |
| 24 October to 6 November | 70 | 118 | The second jump, then a flat line |
Week one (17 to 23 October). The report was empty for the first two days. On 19 October two addresses appeared, both of them not indexed. On 20 October the picture changed in a single day: 65 pages were indexed and 58 addresses were listed as not indexed. Impressions for the week: zero.
Week two (24 to 30 October). On 24 October the indexed pages went from 65 to 70, and the addresses not indexed from 58 to 118. The first impressions arrived that day too; the week closed with 12.
Week three (31 October to 6 November). Neither number moved at all. The week brought 2 impressions.
In short, the movement happened on two days, 20 and 24 October. For the fourteen days that followed, the report showed the same two numbers.
In the second jump, the indexed pages rose by only 5 while the addresses not indexed rose by 60. That is enough of a picture to cause alarm. So we set the total aside and opened the reasons.
What those 118 addresses were
Search Console groups the addresses it has not indexed by reason. We took the same breakdown on two dates: after the first jump, and at the end of week three.
| Reason in the report | 23 October | 6 November |
|---|---|---|
| Page with redirect | 37 | 79 |
| Alternate page with proper canonical tag | 12 | 20 |
| Discovered - currently not indexed | not listed | 14 |
| Crawled - currently not indexed | 9 | 4 |
| Not found (404) | not listed | 1 |
| Total | 58 | 118 |
On 6 November the server error row stood at zero. Let us read the rows in order.
Page with redirect (79). Addresses that redirect to another address land in this row; Google does not index the redirecting address itself. Because we moved every address of the old site with a 301, we expected this row to grow. As Google recrawls the old addresses it sees the redirect and files the address here. Of the 118 addresses, 79 are in this row, roughly two thirds. For us, this number shows that the redirect map is being read.
Alternate page with proper canonical tag (20). Addresses whose canonical tag points to another page, where that page is indexed. Search Console’s help page says there is nothing to do in this case.
Together these two rows make 99 addresses. So 99 of the 118 were addresses the site keeps out of the index on purpose.
Discovered - currently not indexed (14). Google knows the address but has not crawled it yet. This row did not exist on 23 October. We added new guides to the site at the end of October; it is usual for new addresses to wait here for a while. Even so, we are not assuming which addresses they are; they are on our watch list.
Crawled - currently not indexed (4). Google crawled the page but did not index it. It was 9 on 23 October and fell to 4. We are looking at the remaining four one by one.
Not found (404) (1). A single address. The real finding came from here.
When each reason is normal and when it is a sign worth a look is set out in detail in the table in our technical SEO checklist.
The one real problem: an address we never wrote
The address in the 404 row was this: https://detmore.tr/cdn-cgi/l/email-protection. There is no such page on the site, and no such link in the code we wrote.
The source turned out to be Cloudflare. Cloudflare’s Email Address Obfuscation feature rewrites the HTML on its way to the visitor, to hide the email addresses on a page from address-harvesting bots. According to Cloudflare’s documentation, the feature is enabled by default when an account is set up. Our contact address had been turned into a /cdn-cgi/l/email-protection link in this way. A visitor sees no difference; Googlebot, however, followed the link and got a 404.
On 10 November we turned the feature off. The address is now served on the page as plain text. In exchange we gave up this protection against address-harvesting bots; for an agency’s contact address we found that acceptable. For those who want to keep the feature on, Cloudflare’s own suggestion is to add the line Disallow: /cdn-cgi/ to robots.txt.
A single 404 does not sink a site. What makes this row matter is what it teaches: a layer sitting in front of your site can add links to the page that you did not write, and you will not notice it in your own browser. Without that one-address row in the report, we would not have noticed either.
What has not happened yet: impressions
Three weeks brought 14 impressions in total: 0 in week one, 12 in week two, 2 in week three. For a site with seventy pages in the index that is a very small number, and we are not hiding it.
We also think it is normal. Every address is new, and there are no signals from outside the site yet. Google’s site move documentation says that on a medium-sized site it can take a few weeks for most pages to move, and that rankings may fluctuate in the meantime. Getting into the index is the first step; being shown for a query is separate work, and these three weeks recorded only the first. This data gives us nothing to say about rankings or traffic.
Five lessons for your own launch
- Finish the redirect map before launch. If the “Page with redirect” row is growing, the map is working.
- Set the total aside and open the reasons. 118 said nothing alone; 79, 20, 14, 4 and 1 did.
- Record the breakdown with its date. Had we not kept the 23 October breakdown, we could not have seen the “Crawled” row fall from 9 to 4.
- Open the smallest row too. The real finding sat in a row with one address.
- Look at the layers in front of your site. A CDN, a security plugin or a tag manager may be adding links to the page. If you use Cloudflare, search the example addresses in the 404 row for
cdn-cgi.
Common mistakes
- Trying to bring the not-indexed count down to zero, and requesting indexing for old addresses that redirect.
- Leaving old addresses in the sitemap; the sitemap should carry only the new addresses you want indexed.
- Looking at the impressions of the first three weeks and changing titles or addresses again. Every address change means a new redirect and a new wait.
- Passing over the 404 row without opening it because it is “only one address”.
- Looking only at the chart of totals; the chart shows what happened and leaves out why.
What we will watch next
- The match with the sitemap. The sitemap held 72 addresses on day one; with the guides added at the end of October and on 10 November it reached 88. We will follow how the 70 indexed pages overlap with that list by filtering the report to “All submitted pages”.
- The “Discovered” row. Whether the 14 addresses get crawled.
- The “Crawled” row. Which the remaining 4 addresses are, and whether an important page is among them.
- The 404 row. Whether the row closes after the fix.
- Impressions. How the number changes from week to week. We do not look at it every day; once a week on the same day is enough, as with the Monday report.
In our SEO service the first month goes to the technical audit, and this report is part of it. If you work with an agency, ask it to read this report with you in the weeks after launch; what else to ask is gathered in how to choose an SEO agency.
Common questions
How long does it take for a new site to show up on Google?
In our record, getting into the index took days: 65 pages were indexed three days after the report’s first record, and 70 pages after a week. Being shown in searches moved separately; three weeks brought 14 impressions. Google says a move can take a few weeks on a medium-sized site; that is no guarantee of a timescale, and it varies from site to site.
Is it a problem if the “not indexed” number is higher than the indexed one?
Not in itself. We had 118 addresses against 70 pages, and 99 of the 118 were out of the index because of redirects and canonical tags. The place to look is the reasons: if an important page of yours sits in the “Crawled”, “Discovered” or 404 row, it needs action; old addresses that redirect do not.
Why does /cdn-cgi/l/email-protection show as a 404 in Search Console?
If your site is behind Cloudflare and Email Address Obfuscation is on, the email addresses on your pages are turned into this link. When Googlebot follows the link it can get a 404. There are two ways out: turn the feature off and serve the address as plain text, or, as Cloudflare suggests, close the /cdn-cgi/ path to crawling in robots.txt.
If you have ten minutes today, do this: open the Pages report in Search Console and write every reason under “Not indexed”, with the number beside it, into a table under today’s date. Then open the smallest row and look at the example addresses. When you fill in the same table a week later, you will hold your own site’s record and will not have to guess which number moved.
Sources
This article is general information, not legal, financial or technical advice. Platforms, their rules and the law change over time, so some details may have changed since 10 November 2026 or contain small errors. Before you make a decision for your business, check the current official sources or write to us.