Managing Plans

In many cases, you may have a “plan of attack” in mind for how to best populate a dataset. In order to streamline this process, we’ve created the plan API.

Plans allow you to orchestrate a series of enhancement steps that run consecutively as each step completes. This is useful for complex data enrichment workflows.

Creating Plans

To create a new plan:

from structify.types.enhance_property_param import EnhancePropertyParam
from structify.types.enhance_relationship_param import EnhanceRelationshipParam
from structify.types.plan_param import PlanParam

ENTITY_ID = "Entity-12345678-abcd-efgh-ijkl-987654321"

# Define enhancement steps for the given entity
steps = [
    # Enhance company properties
    EnhancePropertyParam(
        entity_id=ENTITY_ID,
        property_name="business_description"
    ),

    # Find relationships
    EnhanceRelationshipParam(
        entity_id=ENTITY_ID,
        relationship_name="invested",
        allow_extra_entities=True
    ),

    # Multiple parallel steps
    [
        EnhancePropertyParam(
            entity_id=ENTITY_ID,
            property_name="founding_date"
        ),
        EnhancePropertyParam(
            entity_id=ENTITY_ID,
            property_name="location"
        )
    ]
]

# Create the plan
plan_id = client.plan.create(
    dataset="startups",
    plan=PlanParam(steps=steps)
)

Managing Plans

List all your plans:

plans = client.plan.list()
for plan in plans:
    print(f"Plan {plan.id}: {plan.status}")

Pause all running plans for a dataset:

response = client.plan.pause_all(
    dataset="startups"
)
print(f"Paused {response.paused_count} plans")

Resume all paused plans:

response = client.plan.resume_all(
    dataset="startups"
)
print(f"Resumed {response.resumed_count} plans")

Tip

Plans are a powerful way to orchestrate complex data enrichment workflows. You can:

• Run steps sequentially or in parallel

• Enhance both properties and relationships

• Pause and resume execution as needed

Note

When a plan is created, it begins executing immediately. Use pause_all() if you need to temporarily stop execution.