Best AI for AdvisorsWhat is a token?
Short answer
The small chunk of text (often part of a word) that AI models read, write and are priced by.
Definition
AI models don't read words the way we do. They split text into tokens, which are often a word, part of a word or a punctuation mark. Context windows, usage limits and API pricing are all counted in tokens.
As a rough rule of thumb, a page of text is several hundred tokens, but the exact count depends on the model and the language.
Why it matters for advisors
Most advisors on a flat monthly plan never see tokens. They matter when a firm builds its own workflows through an API or a connector, where cost scales with how much text goes in and out, and when a tool caps how much you can upload at once.
They also explain usage limits on busy days: long documents and long chats use more of your allowance.
Token limits also explain why AI tools sometimes stop mid-answer or ask you to split a large file. Breaking a long document into sections and asking about each one usually gives better results than one huge upload.
Example
A firm piloting an automated meeting-summary workflow estimates monthly cost by multiplying average transcript length in tokens by meetings per month and the provider's per-token price.
Watch out for
Token counts aren't the same as word counts, so don't budget an API project on word counts alone.
Related terms
- Context windowHow much text an AI model can take in and consider at once, including your prompt, files and the conversation so far.
- Large language modelThe type of AI model behind chat assistants: trained on huge amounts of text to predict and generate language.
- ConnectorA link between two tools that lets data or actions pass between them, such as a notetaker sending notes to your CRM.
Learn more
Last updated .