Add Category Navigation on /theme/ and /plugins/ that Google can follow more efficiently.

[mtobar]

Current issue is Google can not use a “Search Box” and so it is crawling through the few pages linked from the home of /plugins/ and /themes/ then finding tag categories and then crawling through there. Of note, this navigation does not need to have a lot of share of voice, towards the bottom of the page is ok, but I do think it will help many visitors as well as Google.

(Categories TBD by Todd Robinson in near future)

More info on this here:
​https://docs.google.com/document/d/13YxtyPNIz8798EpZT-AXh1SW2QNLGm9PBhtDJo7O16o/edit?usp=sharing

[netweb]

See ​https://wordpress.slack.com/archives/C9GG77GRJ/p1522706611000049 in #council-ops in Slack

[obenland]

Could the intent of this ticket be satisfied with an XML site map?

The challenge with plugin tags is that a) there are 60k of them and b) the vast majority is not very relevant, so I doubt they're overly useful to link to for users.

[obenland]

/cc @todd-inmotion 👆

[todd-inmotion]

Hi @obenland - my apologies for missing this! We could accomplish this with an XML sitemap but seems like it would be the same problem - 60k unique plugin tags, most not relevant. So is our challenge to figure out how to winnow down the plugin tag list to the relevant ones? Like maybe a list of tags that have X or more plugins that share that tag? Then order the tag by popularity and that is either what goes on the bottom of the page or into the XML feed? X would be a number that makes the list a logical size to consume.

We could manually pick tags as well if you see a pattern that makes sense.

[obenland]

With a sitemap there wouldn't be a need to link to tags, we could just link to every plugin directly.

After I made my comment I tried to find out how hard it would be to add a sitemap, so I activated Jetpack's sitemap feature. At the time only pages were included and it didn't have built-in support for custom post types. It seems like someone added plugin to that list though:

​https://wordpress.org/plugins/sitemap.xml

[todd-inmotion]

After thinking some more on it when replying to the thread - it seems like it would be better to link through to the popular tags because then we can get the tag pages - at least the ones with some good numbers - ranking better as well.

As for that sitemap you linked to there - are you saying that was already there? Or that was newly added from what you did?

Thanks!

[dd32]

Replying to todd-inmotion:

it seems like it would be better to link through to the popular tags because then we can get the tag pages - at least the ones with some good numbers - ranking better as well.

To me it seems pointless to add a sitemap specifically for Tag pages, as all plugins contained within the pages should be indexed (And they link to the tag pages, so Google should know about them already).

Some tags are indexed, for example googling WordPress tag cloud pulls up the tag archive, but a lot of other tags seem to pull up plugins that are far more relevant than the tag itself.

As for that sitemap you linked to there - are you saying that was already there? Or that was newly added from what you did?

The Jetpack Sitemap functionality was enabled by @obenland.
The Jetpack Sitemap functionality specifically does not include Tag/Category archives in the sitemaps, as it's not needed when it's listing all public posts/pages/plugins. It doesn't seem to have any support for changing that behaviour.

[obenland]

@todd-inmotion Just so I understand correctly, you’re not just concerned with plugins not being indexed, you’re also looking for tag pages to rank higher?

For the former a sitemap should do the job, if I’m not mistaken.

With all plugins/themes indexed, search engines will also be more aware of their links to tag pages, so that should help. If you’re looking to add additional links to tag pages from the homepage, can we do that through adding to or replacing existing sections?

[todd-inmotion]

@dd32 - we were talking about having links on ​https://wordpress.org/plugins/ itself to the most popular tags, not using the sitemap.xml to point to the tags. If we do it by popularity of the tag based on its members, it should mostly take care of itself.

@obenland - yes on getting the tag pages better indexed if the tag has a reasonable amount of plugins within the tag. This is a great way to get more traffic while providing the visitor a good place to start evaluating plugins in that category.

Your question "If you’re looking to add additional links to tag pages from the homepage, can we do that through adding to or replacing existing sections?"

Yes, adding a section to the bottom with some type of tag cloud should do it. Do you have something in mind?

[obenland]

It's my understanding that we're talking about two different issues, the indexing of plugins by search engines and the better discoverability for users. Rather than attempting to fix both by linking to popular tags, I think we should address them individually for best results.

IMO the indexing issue would be fixed by a sitemap. It would even address the issue of indexing popular tag pages, based on the amount of plugins that link back to them.

Making plugins more discoverable for users can then be done by adding a tag cloud (as proposed in #2856 though I don't believe it actually helps users), or by adding/changing existing term sections on the home page, like the Featured/Popular/Beta section, which would be my preferred solution. What do you think?

[obenland]

In 7600:

Plugins/Themes: Add CPT to Jetpack sitemaps

See #3540.

[obenland]

Submitted the sitemaps last week and 99.58% of pages have been indexed so far. The remaining should be added over the next few days.

[coffee2code]

Improved discoverability for plugins and themes is certainly something that can be improved and something that we can all agree is desirable. How to go about doing that is not so straightforward. Requests for a tag cloud (here and in #2856) have failed to garner traction (that ticket having been closed as maybelater). And now it has been almost 2 years without any additional discussion.

Therefore, as the last remaining unaddressed aspect of this ticket and discussion, I'm closing this as maybelater. The topic can revisited once someone has more of a plan for how to determine the best way to improve discoverability.