Research

Task type: Research

Use Research when a workflow needs deeper external research with a research provider rather than a simple search result list.

Run behavior

• Side effects: This task is read-only or computation-only by default. It returns data for later steps rather than directly writing records or sending messages.
• Credit usage: This task can consume organization credits or model/research capacity depending on your plan and configuration.

When to use this task

• Research a company before meeting prep.
• Find strategic priorities, recent initiatives, or cited public evidence.
• Gather external context before an AI Prompt summarizes or scores it.

Setup tips

• Ask a focused research question.
• Include known company identifiers like domain and name.
• Use lower temperature for factual research.

Examples

Research account priorities

Ask for the top 3 strategic priorities of {{ steps.1.account.name }} using its domain as grounding context.

Outputs

• Research answer and provider-specific metadata.

Common pitfalls

• Using Research for simple transformations that AI Prompt can handle.
• Asking multiple unrelated questions in one research step.
• Forgetting to summarize research before writing it to a CRM field.

Reference details

These details are generated from the workflow task registry and match the fields available in the builder.

Required fields

• query

Variable-aware fields

Fields that accept variable references ({{ inputs.x }}, {{ steps.N.y }}) from the workflow's variable graph:

• messages
• query
• systemPrompt
• userPrompt

Liquid template fields

Fields that are rendered through Liquid (variables plus filters and control flow):

• messages
• query
• systemPrompt
• userPrompt

Model guidance

Use research models for deep external research; avoid for simple transformations.

Builder guidance

When the Workflow Builder Genie configures this task, it follows this guidance:

Keep research prompts grounded in available workflow variables and explicit source requirements.

Default configuration

New instances of this task start with the following configuration:

{
  "name": "Research",
  "outputKey": "research",
  "taskType": "Research",
  "description": "",
  "provider": "perplexity",
  "messages": [],
  "query": "",
  "systemPrompt": "",
  "userPrompt": "",
  "config": {
    "model": "sonar",
    "temperature": 0.2,
    "maxTokens": 1000
  }
}