Guides
Guides for docs that stay accurate
How small SaaS teams keep a help center true to the product after every release, write the docs customers ask for, and answer from them with a link to the source.
Start here
Reading paths
What is documentation drift?
Catch the articles your last release made wrong, and keep them in step with the code.
How to find what's missing from your help center
Turn the questions your docs could not answer into the articles customers need.
How to launch an AI support chatbot for your SaaS
Answer from your docs with a link to the source, and say so when the docs are silent.
llms.txt guide
Make your product and docs easy for search and answer engines to read and cite.
4 guides
Keep docs current
What is documentation drift?
Documentation drift is when your docs slowly stop matching your product. What causes it, what it costs a small SaaS team, how to spot it, and how to stop it without a docs team.
How to keep your help center up to date
A practical routine for keeping a SaaS help center accurate: what to do on every release, who owns what, how often to review, what to update first, and what to automate.
How to find outdated help articles
Six checks for finding help center articles that no longer match your product, from the changelog and tickets to traffic and the code itself, plus how to decide what to fix first.
A pull request checklist for docs
A pull request checklist that keeps customer docs in step with code: which changes need a docs update, a copyable GitHub template, who writes the update, and how to keep it from becoming a checkbox.
1 guide
Find what's missing
6 guides
Answers customers trust
How to launch an AI support chatbot for your SaaS
How to launch an AI support chatbot that answers from your docs, help center, or knowledge base, cites the source, hands off what it can't answer, and stays accurate after you ship.
How to train an AI chatbot on your documentation
Train an AI chatbot on your docs the way that works: retrieval, not fine tuning. Pick the source, structure pages for retrieval, test on real tickets, handle API docs, and keep it current.
What is an AI support agent?
What an AI support agent is, how it differs from a chatbot, what it should handle first, what it should never handle alone, how to judge one, and how pricing usually works.
Why AI customer support needs citations
Why every AI support answer should link its source, what citations reveal about your docs, where they go wrong, what to do when no source exists, and how to keep cited pages correct.
AI chatbot vs knowledge base: which do you need?
An AI chatbot and a knowledge base do different jobs. When you need articles, when you need a chatbot, why a chatbot without a knowledge base is a liability, and how to run both.
Help center vs knowledge base vs docs site
The difference between a help center, a knowledge base, and a docs site, who each is for, which one a small SaaS needs first, and how they fit together.
1 guide
AI discoverability
Questions
FAQ
Where should I start?
If your help center goes stale after releases, start with keeping docs current. If customers ask things your docs don't cover, start with finding what's missing. If you are adding an AI assistant, start with the AI support chatbot guide.
Who are these guides for?
Founders and small B2B SaaS teams who answer support themselves and don't have a docs team or a technical writer.
Do I need usedocs to use these guides?
No. Every guide works with any help center or docs tool. Each one ends with how usedocs handles the same job, if you want it done for you.