Glossary

Natural language search

The difference between remembering the exact subject line and remembering roughly what happened.

In short

Natural language search is a way of searching in which you describe what you're looking for in ordinary words, such as "the pricing deck Sam sent before the offsite," and the system interprets the meaning rather than matching exact keywords. In email, natural language search usually combines an AI model that understands the request with a conventional search over senders, dates, and content.

Keyword search finds messages containing the words you type. Gmail's operators, like from:sam has:attachment after:2026/03/01, make keyword search precise, but you have to know the operators and remember the right words. Natural language search accepts the fuzzy version and works out the filters and synonyms for you. "Deck" can match a file called Q3_pricing_v4.pptx.

You typeKeyword search looks forNatural language search looks for
invoice from the designer last monthThose exact wordsMessages from your designer, in the past month, with an invoice attached or mentioned
when is the kickoffThe words "when" and "kickoff"The most recent message confirming a kickoff date

How natural language search works

Most systems use a large language model to turn your request into structured filters and search terms, run the search, and then read the top results to pick or summarize the answer. That last step is a form of retrieval-augmented generation.

Synced searches your connected email and files for context while it works on a message, so a request for "the signed SOW" turns into the right file attached to a draft. Doing this by hand across tools is covered in how to search email and files together.

Get started

Get early access to Synced.

Join the private beta. Synced drafts the replies, finds the files, and proposes the meeting times. You approve.

Get Early AccessFree plan. No credit card.