Aline's AI features — reports, Playbooks, redlining, and chat — run on tokens. This article explains what a token is, what actually uses them, and how to keep your usage efficient so your allowance goes as far as possible.
Overview
In this article you'll:
Learn what a token is
See exactly which actions consume tokens — and which are free
Check your usage against your allowance
Apply three habits that keep usage low
Know how to increase your allowance if your team's usage grows
What Is a Token?
A token is a unit of AI processing. When Aline's AI reads a document or generates output, it consumes tokens in proportion to how much text it processes. Your account comes with a token allowance, and you can see how much you've used against that allowance in your account.
What Uses Tokens — and What Doesn't
Tokens are consumed only when the AI processes text. That means:
Uses tokens:
Running a report across your documents
Running a Playbook
AI redlining
Asking questions in AI chat
When you run a document against a Playbook or report, the words in that document count toward usage.
Does not use tokens:
Logging in and browsing your repository
Uploading or organizing documents
Sending a document for signature
💡 In short: you're charged for AI thinking, not for using the platform. Storing and managing contracts is free of token cost.
Checking Your Usage
You can review your token consumption against your yearly allowance at any time in your account. For most teams, normal use stays comfortably within the allowance — it's typically only very large, repeated AI runs across big document sets that consume tokens quickly.
How to Keep Usage Efficient
The biggest driver of token usage is running AI across large numbers of documents. A few habits keep it under control:
Filter before you run. Narrow a report to the contracts you actually need analyzed rather than running the AI across your entire repository. Filtering out irrelevant or expired documents first is the single most effective way to reduce usage.
Test on a small subset. Validate a prompt on a handful of documents before running it across thousands, so you're not spending tokens on a prompt that still needs tuning.
Avoid unnecessary re-runs. Get a prompt right on a sample, then run the full set once.
Need More?
If your team's usage is growing — for example, you're processing a large repository or running frequent portfolio-wide reports — your allowance can be increased. Reach out to your Aline contact to adjust your plan.
💡 Best Practices
Clean up before you run. Deduplicating and marking expired contracts means you stop paying to process dead records.
Prototype prompts in AI chat on a few documents before committing them to a large report.
Run big jobs once, deliberately, rather than iterating at full scale.
📚 Related Articles
Refining AI Prompts for Reports, Playbooks & Chat
❓ Need Help?
Want to know whether a planned run will fit inside your allowance? Click the Support button in Aline — tell us the document count and what you're extracting and we can estimate it with you.