The Salma field notes

Prepare business knowledge for an AI assistant

Better source information makes a better starting point.

Salma, an illustrated AI colleague for your business

Collect the information people actually use

Start with service descriptions, business hours, contact routes and the policies your team explains most often. Prefer clear text with meaningful headings. Remove obsolete copies from the proposed source set, and identify a person who can approve each topic.

Resolve contradictions before adding sources

If one document says a service takes thirty minutes and another says sixty, decide which is current. Separate general policy from customer-specific information. Avoid putting passwords, unrelated personal records or information the assistant should not use into the knowledge collection.

Build a small review set

Write down common questions, paraphrases and a few questions that should go to staff. Include an answer that is absent from the sources. Review the resulting conversations for meaning and next steps. Revisit this set when an important service or policy changes.

FAQ

Questions, answered

Should every file be uploaded?

No. Select relevant approved sources. A smaller maintained collection can be more useful than a large unreviewed archive.

Does a document update immediately change every answer?

Do not assume that. Review the supported source-update process and verify important changes in your configuration.

Keep exploring

A closer look

An AI chatbot that knows your business

Give Salma selected business documents and website information so customer answers reflect your services, policies and working hours. Explore knowledge support.

A conversation, not a checkout

Let’s talk about your working day.

Bring your questions. We’ll show the relevant workflows and explore the right setup together.

Request a demo