ivanagas
commented
1 year ago Looks good, my thoughts:
- By organize all existing content, you mean tag into appropriate category or something else?
- I think some startup categories could be amalgamated
- Management, culture, people, founders, operations could be rolled into half as many categories. Maybe culture and leadership.
- Sales and CS should merge
- Same with product engineer
- Product analytics and metrics
- Roadmaps, talking to users could be one
- understanding users can be axed and rolled into talking to users and product analytics.
- Maybe monitoring instead of session replay.
- What is the posts page supposed to rank for? I think its fine if it doesn't rank as long as individual blogs and categories rank better.
- +1 to no category limit
- +1 to line limit
- I think sticky jump links are less important than we think they are, but willing to change my mind if there is data.
- Should we ask YC if they have any advice on creating a content hub?
andyvan-ph
Lior539
corywatilo
joethreepwood
posthog-contributions-bot[bot]
The goal of this issue is:
Decision date: By Sept 6, preferably Desired shipping date: End of September
Background
The Posts view is somewhat inspired YC’s content library, which is a searchable archive of all its startup advice.
As per Cory’s original issue https://github.com/PostHog/posthog.com/issues/6253, the short term goals are:
Long-term ambitions include:
My proposal for content hubs
The original vision, briefly outlined here, was to have one hub covering all topics.
Instead, I’m proposing two content hubs: one for articles about running a startup, and a second just on products / product engineering.
Why two hubs? Because:
While some founders are product engineers, not all product engineers are founders. They aren’t interested in reading about how to do finance, or legal stuff in a startup.
If we had one hub for all this content, the “product” category would be much larger than all the others, which would be weird.
While product engineering content is higher priority for us, our content about startup life / how we work is popular, so it needs to be easy to find.
As per the original plan, the hubs would be a mixture of our own original content, and curated third-party articles, books, newsletters, etc. we recommend.
Process for creating the content hubs
Once we’re ready technically, I propose we do the following:
Here are some drafts lists of categories. They’re not set in stone.
SEO blockers
In hindsight, there are some issues with the original implementation of Posts I should have picked up on earlier. I was too focused on design feedback to study the detail, which is on me.
Here’s a summary from an SEO POV:
URL Structure: Currently articles in the hub have the post type and category in the URL – e.g.
/library/design/article-slug. This creates long URLs, which aren't ideal for SEO. It also means redirecting a lot of existing content (again, not ideal) and forces more in future if we choose to re-org our content. Ideally, URLs should stay as they are, but if they must change then using/posts/article-slugfor most content would be preferable. It also makes it easier to track in analytics.Search indexing: Related to the above, the current posts view is not SEO-friendly at all, so we can’t move our existing blog content until we either make it Google indexable, or come up with some other solution. Making Posts SEO-friendly would probably mean making the filtering less dynamic, which kind of defeats the point in having it, but there are a couple of other solutions we could consider. I won’t detail them all here for the sake of brevity!
One category limit: Because we’re generating filters based on the folder structure, articles can only exist in one category. There are numerous instances where articles could / should exist in more than one filter. Something like the topic tagging on community questions would work if we don't want to rely on front matter.
No categories on posts: Unlike the old blog post template, we don’t have links to the relevant tags / categories on posts. This is less critical than the above issues, but still somewhat important. Google likes this stuff to help understand what articles are about, and it helps Google index our website.
Author names on posts: Same goes for author names, which are currently limited to the sidebar. Google really does care who wrote an article, so it’s important it’s clear and obvious. Would be ace if author names linked to their profile, and / or a filter of posts they’ve written, but that’s more of a nice-to-have.
Static URLs for hubs: If these are to be hubs, we need something we can actually link to. I think it would be better if the filters weren’t hidden in a dropdown, too, but that’s less critical.
Nice to haves
These aren’t blockers, but they are things we should improve when we can.
Post readability: Imo, the line width on posts now is way too long. It makes articles tiring to read, especially longer reads. At full width, lines are ~130 characters. The ideal for the web is in the 70 to 85 character range smashingmagazine.com/2014/09/balancing-line-length-font-size-responsive-web-design/. This also makes full-width images scale super-large compared to the text, which is jarring.
Sticky jump links: Eli added the ‘On this page’ box based on feedback, but it would be preferable is this was sticky so it followed you down the page. I suspect the answer to this and the above issue is to switch a three-column layout. Would be preferable if it ignored H4 headings as well, just so they don’t get too long.
Pinning posts: With the current feed, high-value content is quickly hidden by changelog posts etc. Pinning would be useful to keep things visible, especially for announcements and hero content.