What changed, and how long adoption takes

Key findings

• 131 terms moved up a usage band across three archived months, and 5 moved down.
• Two terms crossed a band boundary and crossed back, so I count them as unstable, not as movement.
• All 87 superseded terms have a documented replacement, and 12 of them are still widely used.
• 22 terms sit on 100,000 websites or more before Schema.org has ratified them.
• Not one of the 877 terms with a datable invention has reached Google's top band.
• A term already on 1,000 websites or more was about 1.6 times more likely to climb than one below that.

131 terms moved up a band, 5 moved down

July 2026 · six ranges, no exact numbersGoogle and the Schema.org community, Schema.org usage statistics dataset. Websites in Google's index.

I count three kinds of movement across the archived months. The table lists the terms whose band changed.

Read as growth, 131 up against 5 down would be remarkable. I do not read it as growth, and I would not want you to. Google's counting settles for a while after a term first appears in the file, so some of this rise is the measurement catching up, not markup spreading. customerRemorseReturnLabelSource and purchaseDate crossed a boundary and crossed back, so I count both as unstable, not as movement either way. If one of your terms moved up this quarter, the honest reading is that it might have.

Three months of band data cannot separate real adoption from settling; the method section explains why.

All 87 replaced terms have a documented successor

The table lists every superseded term in Google's file and the replacement Schema.org documents for it.

12 of them are still on ten thousand websites or more, so this is not a list of terms nobody ever used. Two of the replacements have themselves been replaced, because area points at serviceArea, which itself points at areaServed, and seasons points at season, which itself points at containsSeason. Chains surprised me. I expected a replacement to be final. Where a replacement has its own row here, the table links to it, so you can follow a chain to its end instead of adopting a term that is also on its way out. If a term of yours is in this table, the end of its chain is the term to move to, not the next link.

22 terms are widely used before ratification

The table lists the terms in Schema.org's pending section that are already on 100,000 websites or more.

Most of them are retail, specifically returns and product variants, the same signal the recommended gap report finds from the other direction. Wide use makes a pending term a reasonable choice, not an exotic one. Pending still means the wording is not settled. So if you adopt one of these, I would expect to revisit it.

Schema.org warns that pending terms "are subject to change and should be used with caution", and the method section quotes its full description. You can read it at Schema.org's own documentation for the pending section.

All 51 top-band terms were already published by May 2015

July 2026 · six ranges, no exact numbersGoogle and the Schema.org community, Schema.org usage statistics dataset. Websites in Google's index.

Below I count all 2,487 real terms across Google's six usage bands. The table splits each band by whether the term's invention can be dated.

Not one of the 877 datable terms has climbed into the top band, so the terms the whole web uses were all already in the vocabulary by release 2.0 in May 2015. New vocabulary enters at the bottom. The next section measures how slowly it leaves. Slower than I expected.

Only two of the 877 datable terms have passed a million websites

11 August 2026 · datesSchema.org dated release snapshots. Not a measurement of the web.

Below I count the 877 datable terms by the year Schema.org invented them.

1,610 more terms are excluded because they were already present in release 2.0, which is the floor of my archive, not their true age; the method section carries the full rule.

The same terms appear again by invention year and by the band each one sits in today, so you can see how far any year's terms have climbed.

Not one datable term has reached the top band. Only two have passed a million websites. I can date about a third of the vocabulary here, 877 of 2,487 real terms. The undated two thirds is the shape of the archive, not a fact about adoption. The consequence for a term you adopt is plain. If it was invented in the last few years, the company you are in is small. It will stay small for a while.

Terms already in use were 1.6 times more likely to climb

July 2026 · six ranges, no exact numbersGoogle and the Schema.org community, Schema.org usage statistics dataset. Websites in Google's index.

Below I compare how often terms climbed a band, grouped by where they started.

6.63% of terms that started between 1,000 and 10,000,000 websites climbed a band, against 4.06% of terms that started below 1,000. Adoption compounds. A term with existing implementations has examples to copy, documentation written about it, and tooling that already emits it. This is the finding I trust most on the page, even with the limit the caveat names. A term you can find examples of is a term you can implement in an afternoon, and that alone explains most of the gap.

Part of this gap is the shape of Google's bands, not a difference in adoption; the method section explains why the two cannot be told apart.