The Local Business Lead Scoring Framework

Every local business leaves a public footprint. Most agencies ignore it or evaluate it manually, one business at a time. That does not scale.

Here is a scoring framework that can be automated. It evaluates five public signals, assigns a composite score, and generates a personalized outreach draft. The framework was built for an MCP-powered agent that pulls Google Maps data, but the logic works with any data source.

The five factors

Each factor contributes to a lead score from 1 to 10. Higher scores mean higher lead potential. The scoring is intentionally inverted in places where absence creates opportunity.

1. Website presence (Column 6: website)

No website is the strongest signal. A business without a website needs one. A business with a website needs it to perform. Either way, the conversation starts here.

Scoring: no website detected scores highest. A placeholder or social-only link scores medium. A proper domain scores low unless the site is clearly broken.

2. Review quality (Columns 10 through 15: reviewsCount and reviewsDistribution)

Raw review count matters less than review distribution. A 4.2 rating with 800 reviews where 12 percent are one-star tells a different story than a 4.8 with 50 reviews. High volume plus high negative percentage is the strongest signal. That business has customers but is bleeding reputation.

Scoring: high one-star percentage relative to total count scores highest. Low total count with clean ratings scores low. Unrated businesses are wildcards.

3. Image count (Column 16: imagesCount)

Three images or fewer signals a business that is not investing in its visual presence. This is a photography upsell, a branding conversation, or both. Businesses with zero images are the highest priority.

Scoring: 0-3 images scores highest. 4-10 scores medium. 11-plus scores low.

4. Social media presence (Columns 19 through 24)

Missing accounts create openings. A business on Instagram but not TikTok has a platform gap. A business with no social accounts at all has a visibility problem. The scoring weights missing accounts more heavily than present ones because the offer is clearer.

Scoring: zero social accounts scores highest. Partial coverage scores medium. Full coverage across all platforms scores low.

5. Google Business Profile claimed status (Column 8: claimThisBusiness)

An unclaimed profile is a business that has not taken basic ownership of its Google presence. This is the simplest fix with the highest ROI. It also signals a business that may not know what else it is missing.

Scoring: unclaimed scores highest. Claimed scores low.

Composite scoring logic

The composite lead score is a weighted average:

leadScore = (websiteScore * 0.25) + (reviewScore * 0.30) + (imageScore * 0.15) + (socialScore * 0.20) + (claimedScore * 0.10)

Review quality and website presence carry the most weight. Reviews because they represent existing customer volume and visible reputation risk. Website because it is the highest-value upsell. Image count and social media are multipliers that surface additional need. Claimed status is a quick win indicator that tips borderline leads.

Draft email construction

The personalized email follows a strict format to keep it under 500 characters and comma-free:

• Subject line names the specific gap
• Body opens with the business name and the observation
• References a concrete metric from their data
• Offers a relevant service with a directional outcome
• Closes with a low-friction ask

Example for a restaurant with no website and low review count:

Subject: Helping Redwood Grill Get Found Online

Hi Redwood Grill team I'm [Name] from [Agency]. I noticed your business has no website and only 12 reviews despite a strong 4.6 rating. We help local restaurants build a web presence and grow their reviews. Would you be open to a quick chat about what that could look like for Redwood Grill?

The email never mentions features. It mentions what the business is missing and what that costs them. The offer follows from the data.

How to automate this

The full automation runs on three components:

1. An MCP server that queries Google Maps for businesses by location and category (Apify's Google Maps scraper exposed as an MCP tool)
2. An AI agent with this scoring framework as its system prompt, running inside n8n
3. A Google Drive integration that saves scored leads as CSVs

The workflow: user provides location and category, agent queries Google Maps via MCP, receives structured data for each business, scores each lead against the five factors, generates draft emails, and saves everything to a spreadsheet.

Setup time is about an hour. Running cost is a few dollars per month in API credits. The agent does in 60 seconds what a VA would take an afternoon to do.

What this framework rewards

Businesses that are close to being great but not quite there. A 4.5 rating with 11 percent one-star reviews. A busy restaurant with no TikTok. A dental practice with three photos and no website.

These are the businesses that know they have a problem but do not know it is fixable. The data tells the story. The email opens the conversation. Everything after that is sales, not automation.