Build a HubSpot Workflow That Cleans Your CRM Data While You Sleep

Dirty data does not announce itself. At 100 to 200 seats it shows up as a personalisation token printing "Dear jOHN", a segment that misses a third of the people it should have caught, and a report that undercounts because "Chief Executive Officer" and "chief executive officer" got counted as two different job titles. Then reps stop trusting the CRM enough to use it properly, which is the expensive part.

There is a compliance angle too. Australian Privacy Principle 10, under the Privacy Act 1988, says you have to take reasonable steps to make sure the personal information you hold is accurate, up to date and complete. A workflow that standardises names, emails and phone numbers as they arrive is one of those reasonable steps, and one you can point at if anyone asks. The OAIC's guidance on the Australian Privacy Principles sets out what APP 10 covers.

Steps to Set It Up

Step 1: Map the Fields That Cause the Biggest Headaches

Start with the usual suspects:

• First name/Last name
• Email address
• Job title
• Phone number
• City/Company name
• Industry/Department

Spot patterns like all-caps emails, inconsistent job title casing, or phone numbers full of symbols. You know, the ones that look like someone smashed the keyboard with their face.

While you are in there, note any numbers sitting in text fields. That is a separate job, and converting text properties into number properties is the fix for it.

Step 2: Build Your "Master Data Cleanup" Workflow

1. Go to Automation → Workflows.
2. Create a Contact-based workflow from scratch.
3. Name it something like: Data Hygiene, Standardisation Engine. (Or "The Marie Kondo of CRMs" if you're feeling whimsical.)

This will be the single workflow that keeps everything tidy.

Step 3: Add Smart Enrolment Triggers

Trigger on:

• Property is known → choose each field you want to fix.

Turn on:

• Re-enrol when any of these fields change.

That means any new update, a form fill, a manual edit, an import, gets automatically cleaned. It's like having a spell-check that actually works, but for your entire database.

Step 4: Use "Format Data" Actions to Auto-Correct Fields

This is where the magic happens. Add a Format data action for each field you want to improve.

Examples:

First Name:

• Action: Format data
• Property: First name
• Format: Capitalise first letter of each word
• Result: "jOHN" becomes "John" (as nature intended)

Email:

• Action: Format data
• Property: Email
• Format: Convert to lowercase
• Result: "[email protected]" becomes "[email protected]" (much less shouty)

If email is the only field giving you grief, there is a simpler Ops Hub workflow for cleaning up email data that does the job on its own.

Job Title:

• Action: Format data
• Property: Job title
• Format: Capitalise first letter of each word
• Result: "chief executive officer" becomes "Chief Executive Officer" (proper respect for the C-suite)

Phone Number:

• Action: Format data
• Property: Phone number
• Format: Remove non-numeric characters
• Result: "(555)-123-4567" becomes "5551234567" (clean and consistent)

Stripping symbols gets you consistency. If you would rather the numbers came back in a proper local format, automatically formatting phone numbers goes a step further.

You're essentially building a real-time data polishing machine. (Shinier than a freshly waxed sports car, but for spreadsheets.)

Step 5: Test on a Handful of Messy Contacts

Manually enrol test contacts with:

• all caps emails
• lowercase job titles
• unformatted phone numbers

Check the before/after values to make sure everything behaves as expected. Trust, but verify, especially with automation.

Step 6: Turn It On and Let It Run Continuously

Once published, this workflow becomes your silent guardian. The hero your CRM deserves, but not the one it knew it needed.

Every time a property changes, the workflow automatically fixes it in seconds.

Step 7: Build Your Data Health Dashboard

Make improvements visible and track progress as your workflows clean up your CRM. (Because if you can't measure it, did it even happen?)

1. Set up workflows to flag issues

Because HubSpot lists can't directly detect formatting problems, use workflows to create flag properties for each issue:

Create custom contact properties. Examples:

• "First Name Capitalised?" (Yes/No)
• "Email Lowercase?" (Yes/No)
• "Job Title Proper Case?" (Yes/No)
• "Phone Valid?" (Yes/No)
• "Lifecycle Stage Missing?" (Yes/No)

Build a workflow for each property. The logic checks the relevant property:

• First Name Capitalised? → Check if the first letter is uppercase. Flag "No" if not.
• Email Lowercase? → Flag "No" if email contains uppercase letters.
• Job Title Proper Case? → Flag "No" if job title contains lowercase letters where it shouldn't.
• Phone Valid? → Flag "No" if phone contains symbols.
• Lifecycle Stage Missing? → Flag "Yes" if empty.

2. Build lists based on flagged properties

1. Navigate to Contacts > Lists > Create list
2. Build one active list per dirty data condition, filtering for contacts where the flag property = "Yes" or "No" as appropriate
3. Name lists clearly (e.g., "First Name Not Capitalised") so they're easy to track in reports

3. Turn lists into reports

1. Go to Reports > Reports > Create custom report
2. Choose Single object report > Contacts
3. Filter by List membership to include only contacts flagged for each condition

Metrics to track:

• Current number of records with issues
• Trendlines showing improvements as workflows clean the data

4. Build a Data Health Dashboard

1. Go to Reports > Dashboards > Create dashboard
2. Name it "Data Health Monitor" (or "The Clean Desk Award" if you're feeling cheeky)

Add the following:

• Top data errors: which issues are most common
• Improvement over time: trendlines showing workflow impact
• Workflow-completed actions: how many records have been fixed
• Scorecards for clean vs. dirty data: the one number that tells you how the CRM is tracking

Step 8: Review and Iterate Weekly

• As new errors appear, add new formatting rules.
• As old issues disappear, retire rules you no longer need.

Your CRM gradually becomes cleaner, more standardised, and easier to trust. (Like a garden that weeds itself; the dream.)

Wrapping Up

You've just built a self-cleaning CRM that fixes data problems in real time, as they arrive. No more "jOHN sMITH." No more ALL CAPS EMAILS. Just clean, professional, trustworthy data that makes your marketing and sales teams actually want to use the CRM.

You build it once and it keeps running. Expect to map in the odd new field as the portal grows, which is what Step 8 is for, and you have a record of the reasonable steps APP 10 asks for while you are at it.

Need help building your data hygiene engine? Have a look at how NBH sets up RevOps and HubSpot for growing teams, or get in touch. We promise not to judge your current data situation. We've seen worse (probably).