HubSpot Data Quality: Rules That Keep It Clean

Where does bad data come from?

From a handful of entry points: reps typing records by hand, imports from spreadsheets and events, integrations writing to the same fields, forms with free-text fields, and properties that nobody owns. Each needs its own control, so start by listing which entry points write to your most important fields.

What validation can you set in HubSpot?

Validation rules on single-line and multi-line text, number, phone and date properties, on all plans. Text can be limited by length, characters, case and spaces; numbers by range and decimals; dates to the past, the future, a range or weekdays. Regular-expression rules need Professional or Enterprise.

When we checked HubSpot's documentation in October 2026, rules applied when editing records, in imports and in forms built with the updated editor, but not in workflows, chatflows, meeting pages or legacy forms, and they did not change existing values. That makes them a good fit for manual entry and imports, while workflows and integrations need their own checks.

• Use dropdowns instead of free text wherever the values are known: industry, region, lead source.
• Validate formats that reports depend on, such as IDs, postcodes and phone numbers.
• Keep rules simple enough that reps understand the error message.

Which fields should be required?

Only the ones a process depends on. Requiring email, company, owner and lifecycle stage on contacts, and amount, close date and pipeline on deals, covers most reporting. Requiring ten fields teaches reps to type anything. Required properties on the create form need Starter or above.

For deals, add stage-specific requirements with conditional stage properties, as described in our deal stages guide. For IDs from other systems, use unique value properties, which HubSpot allows up to ten of per object, so the same customer cannot be created twice from an integration or import.

What does HubSpot's data quality tool do?

It shows where the data is already going wrong: duplicate records, formatting issues such as names in capitals or badly formatted phone numbers, and property insights that flag empty, unused or duplicate properties. On Data Hub Professional and Enterprise it can also fix formatting automatically, manage duplicates in bulk and email a weekly digest.

When we checked HubSpot's documentation in October 2026:

• Starter and above: the data quality tool and its property insights. Some formatting-issue features were still a beta on Starter.
• Professional and above: duplicate review one pair at a time, and regular-expression validation.
• Data Hub Professional and Enterprise: bulk duplicate management, automatic formatting rules, the "Format data" workflow action, anomaly alerts and the weekly data quality digest.
• Enterprise: view and edit restrictions per property, which HubSpot says should not be treated as a security control.

Who should own HubSpot data quality?

One named person, usually the HubSpot admin or RevOps lead, with a short rulebook: who can create properties, how imports are approved, which system owns each synced field, and a monthly review of the data quality tool. Without an owner, every rule above slowly erodes.

1. Review the duplicates and formatting issues in the data quality tool, or the weekly digest on Data Hub Professional.
2. Merge and fix what it finds, and note which entry point caused each problem.
3. Check property insights for empty, unused or duplicated properties, and retire or merge them.
4. Audit the entry points that caused problems: an import template, a form field, an integration mapping.
5. Update the rules so the same problem cannot come in the same way again. Our governance checklist covers the rest.

Data quality checklist