Use Hidden Helper Properties to Normalise Data in HubSpot

Overview

This hack introduces a hidden normalisation layer inside HubSpot using helper properties and workflows.

Instead of relying on inconsistent user input, you standardise data behind the scenes, making automation, reporting, and lifecycle logic reliable and predictable.

---

The Problem

Uncontrolled data entry in HubSpot leads to:

• Inconsistent values
  • “Paid Search” vs “paid search” vs “PPC”
• Duplicate properties representing the same concept
• Workflow enrollment failures due to unexpected values
• Broken lists and unreliable reports

---

The Hack

Create hidden helper properties that:

• Convert messy input into a controlled set of values
• Act as the single source of truth for automation and reporting
• Sit between raw data and business logic

---

Why It Works

• HubSpot workflows require exact value matches
• There is no native normalisation layer
• This approach creates one using:
  • Properties (data model)
  • Workflows (mapping logic)

Result:

---

When to Use This

Apply this hack to properties that are:

• Free-text or loosely controlled
• Populated from multiple sources (forms, imports, integrations)
• Used in:
  • Workflows
  • Lists
  • Reports
  • Lifecycle or routing logic

Common candidates:

• Lead Source
• Industry
• Product Interest
• Region / Territory
• Use Case

---

Step-by-Step Implementation

1. Define Your Clean Output Model

Decide what “good” looks like.

Example: Lead Source

Raw inputs:

• Google
• Google Ads
• PPC
• Adwords

Normalised values:

• Paid Search
• Organic Search
• Email
• Referral
• Direct
• Other

Keep this list:

• Short
• Stable
• Business-relevant

---

2. Create the Helper Property

• Go to Settings → Properties
• Select the correct object (e.g., Contacts)
• Create a new property:

Configuration

• Label: Normalized Lead Source
• Internal name: normalized_lead_source
• Field type: Dropdown select
• Options: your normalised values
• Visibility: Hidden

---

3. Build the Normalisation Workflow

• Go to Automation → Workflows
• Create a workflow (same object as property)
• Start from scratch

---

4. Set Enrollment Criteria

Use:

• Original property is known
• AND helper property is unknown

This prevents unnecessary overwrites during initial setup.

---

5. Add Mapping Logic

Use If/Then branches to map inputs → outputs.

Example:

• If Lead Source is:
  • Google
  • Google Ads
  • PPC
  • Adwords
    → Set normalized_lead_source = Paid Search
• If Lead Source is:
  • SEO
  • Blog
    → Set normalized_lead_source = Organic Search
• Else → Other

Each branch ends with Set property value

---

6. Handle Edge Cases

Always account for unknowns:

• Add a fallback value: Other or Unknown
• Optionally track issues:

---

7. Test Before Activating

• Manually enroll test records
• Validate:
  • Correct mappings
  • No overwrites
  • No re-enrollment loops

Only activate when results are consistent.

---

8. Switch to the Helper Property

Once stable:

• Update workflows to use the helper property
• Update lists and filters
• Update reports

The original field becomes:

Input only — not logic-critical

---

Best Practices

Naming Convention

Use consistent prefixes:

---

Do Not Expose Helper Properties

Keep them out of:

• Forms
• Default record views
• User-editable contexts

---

Allow Reprocessing (Advanced)

To keep data accurate over time:

• Enable re-enrollment
• Allow workflow to update the helper property if the source changes

---

Monitor Data Quality

Add supporting properties:

• normalization_status
• needs_data_review

Flag anything that doesn’t map cleanly.

---

Limitations

This approach may not scale well if:

• You have hundreds of possible values
• Inputs require fuzzy matching or NLP
• Data should be standardised upstream (ETL, integrations)