Sitemap lastmod uses last publish date instead of last page edit date

[johnsicili]

Sitemap lastmod uses last publish date instead of last page edit date

[istarkov]

Technically even if content of the page is not changed but scripts are, css is changed for the whole site too. We are not cms so almost every publish has effect on the whole app. Its even without slots etc. Internally we have nothing now to detect the latest change of the page too, I would postpone this after collaboration data structures where at least we will have some kind of time for every change

[kof]

@istarkov true, lets see how webflow treats this

[kof]

We need to learn if this date should be updated when anything changes or when only content changes and then think about all kinds of content possibly changing: text, variables, content that is fetched remotely. It is a challenging thing to solve indeed.

[kof]

@johnsicili is the date correct in case of dynamic pages where content comes from CMS?

[istarkov]

Based on my understanding, the primary reason for considering this field is to manage the "crawl budget" effectively. There is a theory that Google imposes certain constraints on how often it will recrawl your site. For websites with 100-500 pages, this is typically a small number, and worrying about the crawl budget may not be necessary. However, for larger sites, especially those without a Content Management System (CMS), managing this can be challenging. BTW do we think we will really have cms less 500+ pages sites

[kof]

True, on a static site crowl budget isn't gonna be a problem on a dynamic cms-based one, date will be correct based on content changes.

[kof]

Thinking about it if CMS based sites report correct dates, I think we should just close this issue and not bother at all.

[johnsicili]

CMS properly reports lastmod because they are bound to the external value. I'm pinging an SEO group for further clarification on the subject.

[johnsicili]

BTW, Webflow does not use lastmod.

https://discourse.webflow.com/t/add-and-automatically-update-lastmod-tag-in-sitemap/225322

[johnsicili]

I asked John Mueller of Google the following question:

TL;DR – If Google determines lastmod isn’t “consistently and verifiably accurate”, do they treat all URLs in the sitemap this way? Or is it conditional? If I have a batch of URLs that have accurate lastmod dates and a batch that have not so accurate, is it best to use two separate sitemaps?

He responded:

I wouldn't see the lastmod date as a magic bullet. Use it where you have something useful to share, skip it otherwise. If you do that in separate sitemap files, that might make it easier for you to maintain. It's also fine to share whatever level of accuracy that you have - it doesn't need to be millisecond-level (it can be by day).

Depending on the site, you might also consider RSS feeds, which have similar recommendations in terms of dates: https://developers.google.com/search/blog/2014/10/best-practices-for-xml-sitemaps-rssatom

---

It doesn't directly answer the question if the entire sitemap's lastmod will be disregarded if it finds a handful of them not accurate.

IMO keeping them at least has upside potential – Google will use them. If there are too many inconsistencies, then they might not use them which likely is the same as not having the date altogether. I would go for two separate sitemaps just in case though.

Another note is that lastmod should only be used if there are meaningful changes.

[kof]

Sounds like we can close the issue?