How to build an investment-grade company scoring system with Make and alternative data

Most company scoring systems rely on static firmographic data or non-transparent machine-learning models. The result is either outdated insights or scores that are impossible to explain.

But what if you could build a time-aware, explainable company scoring system using real-world signals without writing any code?

In this guide, we’ll walk through how to use Make and alternative data to build an investment-grade company scoring workflow using Google Sheets as the output layer.

Why alternative data matters for company scoring

Traditional datasets answer questions like:

• How big is the company?
• Where is it located?
• What industry is it in?

Alternative data answers more important questions:

• Is the company hiring right now?
• Are they investing in leadership and new tools?
• Is something changing internally?

By using signals such as job openings, news events, and technology adoption, you can measure momentum instead of static attributes.

Real-world use cases: Who it is for

This type of scoring system can be used for:

• Investment screening
• Corp dev target prioritization
• Sales and GTM account scoring
• Partner evaluation
• Market monitoring

What makes a scoring system “investment-grade”

Before building the workflow, let’s define the requirements.

An investment-grade scoring system should be:

• Time-based - focused on recent activity rather than snapshots
• Explainable - every score can be traced back to signals
• Composable - signals can be added or removed easily
• Automated - no manual updates or scripts

Overview: the Make workflow architecture

At a high level, the scenario looks like this:

Each company is processed independently, and all signals are written back once per run.

Step 1: Set up your company list in Google Sheets

Create a Google Sheet with one row per company:

Example columns:

• domain
• jobs_90d
• jobs_30d
• senior_roles_90d
• score

This sheet will act as both your input and output layer.

Step 2: Create a new Make scenario

1. Log in to your Make account
2. Click Create a new scenario
3. Add Google Sheets → Get rows as the first module
4. Select your company list sheet

Step 3: Iterate through companies

Add:

• Tools → Iterator

Map the rows from Google Sheets.

From this point on, every step runs once per company domain.

Step 4: Count job openings in the last 90 days

Now let’s add the first and strongest signal: hiring activity.

PredictLeads → List Job Openings

Configure:

• Domain = current row domain
• first_seen_at_from = now - 90 days

This returns one bundle per job opening.

Tools → Array Aggregator

Collapse all job bundles into a single array.

Tools → Set variable

Create:

jobs_90d = length(aggregated_jobs)

This gives you the total number of new roles opened in the last 90 days.

Step 5: Identify senior hiring activity

Leadership hiring often signals strategic growth.

PredictLeads → List Job Openings (90 days)

Reuse the same 90-day query.

Filter

Keep only roles where:

• seniority ∈ manager, director, head, VP, executive

Array Aggregator + Set variable

Create:

senior_roles_90d = length(aggregated_senior_jobs)

Step 6: (Optional) Add short-term momentum

Repeat the same pattern with a 30-day window:

jobs_30d = job openings first seen in the last 30 days

This allows you to measure hiring acceleration.

Step 7: Update Google Sheets (single write)

At the very end of the scenario:

Add Google Sheets → Update row Match rows by:

• domain

Update:

• jobs_90d
• Jobs_30d
• senior_roles_90d
• last_scored_at

This ensures all metrics land in the same row.

Why does it give you

This approach ensures:

• No race conditions
• No partial updates
• Fully explainable numbers
• Easy debugging
• Easy extension

Each dataset is processed independently, and Make acts as the orchestration layer.

Extending the workflow with more signals

Once the jobs block is stable, you can easily add:

News events

• Count high-confidence events
• Detect negative signals like layoffs

All categories can be found here.

Technology adoption

• Measure modern tech stack usage
• Detect recent tooling changes

Each signal follows the same pattern:

Fetch → Filter → Aggregate → Variable