SEO and AJAX

[hancush]

On the live site, all processing work is done server-side, so the page loads with data already in place, making it perfectly indexable by Google, et al.

In version two, we're transitioning an AJAX approach to retrieve metadata, such as summary statistics, top earners, and any child departments, based on data year. This means data is not in place at page load, which will impact SEO.

[hancush]

N.b., Person pages will display all data on load, at once.

[hancush]

Some initial learning:

• Provided external JavaScript files are indexable and scripts are valid (i.e., not broken), Google will render JavaScript.
• Google indexes both raw and rendered HTML in "waves".
  
• It's advisable to push pushState or replaceState (the JavaScript history API) to update URLs when a JavaScript interaction changes page content. Do not use hash navigation! Google considers it a hack, and any content after the hash is ignored.
• Some instances where dynamic content will not be indexed:
  • JavaScript is broken
  • Content is hidden behind user interaction (Google will only index content displayed on page load)
  • Content takes longer than 5 seconds to load

Links:

• https://developers.google.com/search/docs/guides/javascript-seo-basics
• https://moz.com/beginners-guide-to-seo

[hancush]

Action items here:

• Update employer pages to default to most recent year data is available.
• Build real URLs corresponding to data year via the JavaScript history API.
• Ensure API content loads quickly enough (Improving API efficiency #437).
• Add a description meta tag to entity pages. The result snippet is currently FUGLY.