The channel search box, the tag cloud, the folder list, and the popular-search sidebar are all powered by the same backend: Azure Cognitive Search. Understanding how it indexes your articles makes it easier to predict how they'll be found.
How indexing works
Every time you save an article, OmniCon updates the Azure Search index. The index stores the article's title, description, body text, tags, folder, author, culture, and publish window — everything the search box and the sidebar facets need.
There's usually a small delay (a few seconds) between saving and the article appearing in search results. The article itself is reachable by permalink immediately — only the search index lags.
The Searchable flag
Each article has an Is Searchable toggle. When off:
- The article is excluded from the search index.
- It doesn't appear in keyword search results.
- It doesn't contribute to tag or folder facet counts.
- Related-article recommendations won't include it.
But it's still reachable by direct URL. Good for:
- Landing pages linked only from ads.
- Internal-reference articles you don't want cluttering search.
- Holding pages that exist just to keep a permalink alive.
Sidebar facets
The Tags and Filter by Directories sections in the sidebar come straight from Azure Search facet counts. That means they always reflect what's in the index right now, including the publish-window filter.
If a tag doesn't show up when you expect it to, check:
- Is the article inside its publish window?
- Is it marked Searchable?
- Is it not archived?
- Did you wait a few seconds for the index to catch up?
Popular Search Terms
The third sidebar section — Popular Search Terms — is different. It's not facets; it's a history of what readers have typed into the search box. Every non-empty keyword is recorded and counted. The top 10 show up as clickable suggestions.
This is a useful signal: if "pricing" shows up with a big count, maybe it's time to write an article with that word in the title.
Search query syntax
The search box accepts plain keywords. Behind the scenes, Azure Search tokenizes and ranks results by relevance. No Boolean operators or advanced syntax in v1 — just type what you're looking for.