CRM Cleanup - Contact Level

Last Updated At: Apr 2026

​

Why this playbook exists

CRM data decays over time. People change jobs, companies change names, and profiles get updated. Even if your CRM was perfect when the deal was created, it will drift over months. The fix is consistent re-enrichment. This playbook is designed to refresh HubSpot contacts end-to-end, without wasting credits on data that is already valid.

​

What this playbook assumes

This workflow assumes that for each HubSpot contact:

• the contact record exists in HubSpot
• the contact has an email

If a contact does not have an email at all, this exact playbook may not work as-is, because email is used as a key input to recover LinkedIn and enrich the profile.

​

How to run the playbook

If you have the playbook set up in a grid already:

1. Import contacts into the grid
2. Select all rows
3. Click Run

That is it. The rest is handled by the workflow.

​

Step 1: Pull contacts from HubSpot

1. Go to New grid
2. Select Import records from HubSpot lists
3. Select your HubSpot account
4. Select the list
5. Import into:
   • a new grid, or
   • the current grid if you are using the playbook template

This import gives you HubSpot record IDs, which become the base input for everything else.

​

Step 2: Look up key HubSpot contact fields

Once HubSpot IDs are available, the workflow runs a lookup to pull common contact fields like:

• name
• company name
• company website
• LinkedIn URL (if present)
• email

These are the core fields most teams care about at contact level.

​

Step 3: Validate the existing email

Before spending credits to find a new email, the workflow validates the email already in HubSpot. Why this matters:

• if the current email is valid, there is no need to re-find it
• you can still enrich missing fields like name, company name, LinkedIn URL
• but you avoid paying to “re-enrich” something that is already correct

Only contacts with invalid or low-quality emails flow into the “find new email” step later.

​

Step 4: Find LinkedIn profile URL

LinkedIn profile URL is treated as a high-value key because it unlocks richer, fresher data. The workflow tries two methods together:

1. Reverse lookup waterfall
   • pass the email into multiple providers
   • attempt to return the LinkedIn URL
2. Bit Agent fallback
   • if the waterfall fails, Bit Agent searches to find the LinkedIn profile URL
   • this increases coverage without guessing

​

Step 5: Enrich the person from LinkedIn (only when needed)

The workflow enriches the person to pull fresh, live data from LinkedIn. But it does this conditionally. If key fields like company website, company name are already present, you may not need to enrich the person for those parts. If those fields are missing or empty, the workflow enriches the person to fill them.

​

Step 6: Re-find email only for invalid contacts

For contacts where the email validation step marked the email as bad or invalid:

• the workflow finds a new email
• keeps everything grouped and clean for update

Valid emails do not go through this step.

​

Step 7: Update the HubSpot record

Finally, the workflow pushes updates back to HubSpot:

• action: Update record
• unique identifier: HubSpot record ID
• map whichever fields you want to update
  • example: LinkedIn URL, name, company name, company website, updated email (only when needed)

​

Automate it with scheduled fetch + auto run

Once the playbook works for your dataset, you can make it fully automatic.

​

Option A: Use a HubSpot list as the feeder

This is best when your team manages eligibility through list filters.

1. Set Auto-fetch records on the HubSpot list source
2. Pick a cadence (example: bi-weekly, monthly)

​

Option B: Pull contacts using HubSpot querying

If you want bulk re-enrichment (example: every 6 months), you can skip lists and pull contacts using HubSpot contact querying directly. The workflow still works because it only depends on:

• HubSpot record ID

​

Enable auto run + auto dedup

To make the pipeline hands-off:

• enable Auto run so every new record runs through the workflow
• configure Auto dedup to avoid processing the same contact repeatedly

Whether you want dedup or reprocessing depends on your cadence:

• frequent cadence: dedup helps avoid repeated enrichment
• long cadence (6–12 months): reprocessing may be desirable

​

Need help?

If you want a copy of this playbook, or help adapting it to your HubSpot setup and cadence, reach out on the support channel. The team will help you get it running end-to-end.