> ## Documentation Index
> Fetch the complete documentation index at: https://docs.leadinfo.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Build a daily action list

> Give Sales a short list of companies worth acting on today, each with the best person to contact and the reasons behind the score.

The opportunities endpoint does the ranking, the exclusions and the contact selection for you. This guide adds the two steps that make the list trustworthy: the explanation behind each company and the check for existing relationships.

**1. Pull today's opportunities**

```bash cURL theme={null}
curl --request GET \
  --url "https://data.leadinfo.com/v1/trackers/LI-5A14070B93509/opportunities?limit=10" \
  --header "Authorization: Bearer <token>"
```

Without dates the window is today. Companies are ranked by `engagement_score` over the last 90 days. The default `relationship=prospects` leaves out companies already in a connected CRM or tagged customer, competitor or not-interesting, and AI assistants and shared-network traffic never appear.

Each item has a `company` (a full company row) and a `contact`:

```json theme={null}
{
  "company": { "company_id": 67528790, "name": "Van Beek Ingenieurs B.V.", "score": 67, "engagement_score": 64, "in_crm": false, "...": "..." },
  "contact": {
    "source": "contact",
    "name": "J. de Vries",
    "job_title": "Marketing Manager",
    "linkedin_url": "https://www.linkedin.com/in/...",
    "matched_personas": [14229],
    "has_email": true,
    "email_verified": true
  }
}
```

`source` is `visitor` when the person identified themselves on your website and `contact` when they come from Leadinfo's contact database. Only the first kind is known to have visited.

**2. Explain the top companies**

```bash cURL theme={null}
curl --request GET \
  --url "https://data.leadinfo.com/v1/trackers/LI-5A14070B93509/companies/67528790/score" \
  --header "Authorization: Bearer <token>"
```

The response lists each pillar with the rules matched and missed, the pages viewed, `known` facts and `unknown` gaps. Put the three questions side by side in your hand-off: does the company fit (fit pillar), did it show interest (intent and engagement, with the page topics), and is a next step justified (people pillar, CRM status, tags).

**3. Choose the action**

A practical mapping:

| Signal pattern | Suggested action |
| - | - |
| High fit, pricing or demo pages in the window, nobody identified | Research the account, then a persona contact with a topic-based opener |
| High fit, only generic pages | Awareness: add to a nurture audience, no personal outreach yet |
| Any fit, `in_crm` true or tagged customer or partner | Hand to the owner in Sales; do not prospect |
| Persona-matched identified visitor | Reach out; the person is known to have visited, so a direct follow-up is appropriate |
| `is_shared_network` or `is_ai_agent` | No action; report as traffic noise or as assistant activity |

**4. Refine the person**

When the recommended contact does not fit, list alternatives with `GET .../companies/{company_id}/contacts?persona_id=14229`, or fall back to `dmu=marketing` or `department=leadership` at small companies where titles are informal.

> [!WARNING]
>
> Keep page-level behaviour out of the message you send. Describe the topic the company showed interest in as a possible need, and never imply that the contact personally visited unless they are an identified visitor.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.