Six new sites, one launch day, 106 URLs. Six days later Google had indexed 94 of them.

The line you hear is that a new domain sits in a sandbox for weeks. Sometimes months. Nobody I have read ever put a number on it with a per-URL check, and I happened to have six sites that went live in the same week, so I stopped repeating the line and went to look.

What I had to work with

Six sites. Same stack, same sitemap format, same launch week, different topics, and not a single backlink pointing at any of them. Small ones, between sixteen and nineteen addresses each.

I asked the URL Inspection API about every single address. Not a sample, every one, six days after launch. It answers per URL, which is the whole reason this was worth doing.

Ninety-four of them were indexed.

Do this before you trust the indexing report you already have

Open the Sitemaps report in the API. Look at the indexed field. Then take ten of those same URLs and put them through URL Inspection.

If the two disagree, believe URL Inspection.

Mine disagreed by everything. The Sitemaps API reported zero indexed pages for all six sitemaps, in the same hour that URL Inspection reported ninety-four indexed pages sitting behind them.

I had already built a dashboard on that zero. It said the sites were dead and that indexing was the thing blocking growth. Both conclusions were wrong, and I worked from them for a day before I checked.

The misses are only two kinds

Twelve addresses were not indexed. None of them were blocked by robots, excluded by a tag, redirected, or flagged as duplicates.

Two states only. Either Google had never heard of the address, or Google had it from the sitemap and had not fetched it yet.

Those have different fixes, and neither fix is “wait”.

The crawl arrived in a lump

On one site, fifteen of its sixteen pages carry the same last-crawl date, and the sixteenth is two days later.

I had expected a drip. What it looks like is one visit that took almost everything, then a slow return for the leftovers.

If that holds elsewhere, the sitemap is doing more work in launch week than internal linking is, which is the reverse of the order most launch checklists use. Six sites is not enough to claim it. It is enough to measure properly on the next launch, and measuring costs nothing.

An afternoon I would not spend again

I tried to correlate index share with page count, text volume and average response time. Nothing came out. Six points is too few for a correlation, which I knew while I was doing it, and I did it anyway.

Now the spread, which is the part I keep

Across the whole set that comes to 88.7% in six days, on domains registered the week before that nobody links to.

Same launch week, same stack. One site got all of it. Another got seven tenths.

What I still cannot explain

Site F sits at 71%, and four of its five missing pages were discovered and never fetched. Those four are the only pages in the set that no internal link points at.

Tidy story. I do not fully believe it, because site E has orphan pages too, and Google indexed those.

Could be crawl depth. Could be the day of the week. Could be nothing at all. The next check is the same one across the next six launches, with orphan status written down before the crawl instead of after it.

Questions people ask about this