Importing historical data to train the agent

Plan requirement

Subscription Any plan
Also required AI agents - Advanced
Access Admin

Past conversations can teach the agent how your customers actually ask. Powerful, and it needs a decision about what you feed it.

Import it

  1. Open the AI agent's training settings.
  2. Choose the historical conversations to use.
  3. Start the import.
  4. Review what it derived before acting on it.

What this gives you

The phrasing customers really use, and the frequency of each subject. That is information you cannot get from your own imagination, and it usually contradicts the internal names for things.

It is as valuable for deciding what to document as for the agent itself.

Choose the period deliberately

Recent conversations reflect the current product; old ones describe how things used to work. A year is usually the right window, and less if you have changed something significant.

Old answers were not all correct

Historical exchanges include agents who were mistaken, policies since changed, and workarounds no longer needed. Treat the import as a source of questions rather than a source of answers.

Think about what is in them

Real conversations contain real customer details. Consider what your tickets hold before importing them, and involve whoever is responsible for that decision in your organisation.

Read what came out

The subjects it identified are a report on your support operation. Anything frequent and undocumented is an article to write, and that list is worth more than the training itself.

See also

Was this article helpful?

0

Still stuck?

Our support team will take a look with you.

Comments

0 comments

Article is closed for comments.