Update GHL custom fields from AI conversations
Configure your AI agent to update GoHighLevel custom fields automatically during calls and chats using the update_user_details tool.
Your AI agent can update GoHighLevel custom fields in real-time during conversations. When the agent collects information — budget, timeline, service interest, company size, or any other data point — it writes the value directly to the contact’s custom field in GHL. No manual data entry, no post-call workflows needed. The data is available in your CRM the moment the conversation happens.
Set up custom field updates
- Create custom fields in GHL
In your GoHighLevel account, go to Settings → Custom Fields and create the fields you want the AI to populate. Note the exact field names — the AI agent needs to reference them precisely.
Common fields for AI agents:
- Budget range
- Service interest
- Timeline
- Company size
- Lead qualification score
- Preferred contact method
- Open your agent in the Agent Builder
In Centerfy, go to AI Agents and select the agent you want to configure.
- Enable the update_user_details tool
Go to the Tools & Abilities tab and enable
update_user_details. This tool allows the AI to write to contact custom fields during conversations. - Configure the fields
In the tool configuration, map the fields the AI should update. Specify:
- Field name — the exact GHL custom field name
- Field type — text, number, dropdown, date, etc.
- When to update — describe the conditions in your system prompt
- Update your system prompt
Add instructions telling the AI when and how to update each field. Be specific about what information maps to which field. For example:
“When the caller mentions their budget, update the ‘Budget Range’ custom field with their answer. When they state what service they’re interested in, update the ‘Service Interest’ field. Always confirm the information with the caller before saving it.”
- Save and test
Save the agent and run a test conversation. Mention specific data points (e.g., “My budget is around $5,000”) and verify that the corresponding custom field is updated in the contact’s GHL record.
How it works
During a conversation, the AI:
- Asks qualifying questions or naturally collects information based on the caller’s responses
- Identifies which custom field the information belongs to
- Calls the
update_user_detailstool with the field name and value - The field is updated immediately in the GHL contact record
- The updated data is available in pipelines, workflows, reports, and automations instantly
Best practices
- Use descriptive field names —
budget_rangeis better thanfield_1. The AI references these names, and clear names reduce errors. - Define expected formats — in your system prompt, specify formats (e.g., “Store budget as a number without currency symbols” or “Store dates as YYYY-MM-DD”).
- Limit fields per conversation — don’t try to fill 20 fields in one call. Focus on the 3-5 most important data points for lead qualification.
- Use dropdown fields where possible — constrained fields (dropdown, radio) are more reliable than free text because the AI selects from predefined options.
Use updated fields in workflows
Once the AI updates a custom field, you can trigger workflows based on the new value:
- Pipeline automation — move a contact to a different pipeline stage when their budget field is populated
- Tagging — automatically apply tags based on service interest
- Notifications — alert your sales team when a high-value lead is identified
- Follow-up sequences — trigger different follow-up sequences based on the qualification data collected
Can the AI update standard GHL fields too?
Yes. The update_user_details tool can update both standard fields (name, email, phone, address) and custom fields. Use it for any contact attribute you want the AI to populate.
What happens if the AI writes to a field that doesn't exist?
The update will fail silently. The AI won’t crash or error out, but the data won’t be saved. Always create your custom fields in GHL before referencing them in the agent configuration.
Can the AI overwrite existing field values?
Yes. Each update overwrites the current value. If you want to preserve historical values, use a “notes” field or a separate field for each interaction rather than overwriting the same field repeatedly.
Does this work with both voice and chat agents?
Yes. The update_user_details tool works the same way for both voice agents and chat agents. Any information collected during a conversation can be written to custom fields regardless of the channel.