# Lead Qualification & Enrichment

> Score, enrich and route every inbound lead within seconds

| | |
|---|---|
| **Category** | Sales |
| **Type** | API Pipeline |
| **Difficulty** | Intermediate |
| **Automation** | Fully automated |
| **Runtime** | ~4 min |
| **Est. cost** | $0.18 |
| **Version** | 1.0.0 |

## Purpose

Inbound leads decay fast. This pipeline enriches and scores each one within seconds of creation, so the right rep gets the right lead with context already attached rather than a name and an email.

It is fully automated by design — the judgement lives in the scoring rubric, which a human owns and tunes.

## Problems solved

- Reps spend the first ten minutes of every lead researching the company
- Hot leads sit in a queue behind cold ones
- Scoring is inconsistent between reps
- CRM records stay thin, so reporting is unreliable

## Benefits

- Response time measured in seconds
- Consistent scoring against one rubric
- Reps open a lead with context already there
- CRM data quality improves as a side effect

## Limitations

- Enrichment quality depends entirely on the data provider
- The scoring rubric needs periodic retuning as the ICP shifts
- Enriching personal data has real GDPR implications — check your lawful basis
- Junk in, junk out: bad form data still produces bad scores

## Success metrics

- Median time from lead creation to routed
- Percentage of leads correctly routed first time
- Meeting booked rate versus the manual baseline
- Cost per enriched lead

## Prerequisites

- **Api key** — LLM API key
- **Account** — CRM with webhooks
- **Account** — Enrichment provider
- **Permission** — Lawful basis for enrichment

## AI tools used

- **ChatGPT**

## Steps

### 1. Lead created trigger

CRM webhook fires with the raw form submission.

- _Tool:_ HubSpot
- _Time:_ 1 min

### 2. Normalise and deduplicate

Clean the email and domain, then check whether this contact or account already exists.

- _Time:_ 1 min

> ⚠️ Merge rather than create when a duplicate is found, or reporting drifts immediately

### 3. Enrich from public sources

Pull firmographics: size, industry, funding, tech stack, location.

- _Tool:_ Clearbit
- _Time:_ 1 min

### 4. Score against the ICP

Apply the scoring rubric and produce a score with written justification, so a rep can challenge it.

- _Model:_ GPT-4o
- _Time:_ 1 min

```text
Score this lead against our ideal customer profile.

ICP:
{{icp_definition}}

Lead:
{{enriched_lead}}

Return JSON:
{
  "score": 0-100,
  "tier": "hot|warm|cold",
  "reasoning": "two sentences on why",
  "disqualifiers": ["any hard blockers"],
  "talking_points": ["three specific angles for the first message"]
}

Be strict. A lead that only partially matches is warm, not hot.
```

### 5. Write the briefing note

Produce a short brief the rep reads before their first touch.

- _Model:_ GPT-4o
- _Time:_ 1 min

### 6. Route to owner

Assign by territory, tier and capacity, then notify in Slack with the brief attached.

- _Tool:_ Slack
- _Time:_ 1 min

## FAQ

**Is this GDPR compliant?**

The workflow can be, but compliance depends on your lawful basis for enriching personal data and on your provider's terms. Confirm both before switching it on in the EU or UK.

**What if the scoring is wrong?**

That is why step 4 returns written reasoning. Sample a week of scores against actual outcomes and tune the rubric — the model applies your definition, so a bad score usually means a vague ICP.

**Can it work without a paid enrichment provider?**

Partially. You lose firmographics, which is most of the signal. Domain-only heuristics are noticeably weaker.

---

_Exported from AIToolsay — https://invitationbuddy.com/workflows/lead-qualification-enrichment_