How to match accounts between Salesforce and HubSpot after an acquisition

Salesforce's and HubSpot's duplicate tools each see only their own records. How to match an acquired company's HubSpot data to your Salesforce accounts.

The short answer: Salesforce's duplicate rules and HubSpot's duplicate tools each compare records inside their own system, and the HubSpot-Salesforce integration pairs records only on exact keys such as an email address, so you match the two customer bases outside both CRMs. Export both systems, normalize names and domains, match on several signals at once, and have a person review the uncertain pairs. Record each decision in a crosswalk of Salesforce IDs and HubSpot record IDs, then choose the surviving system and load with those IDs so every record lands on its match.

This post covers the mixed case, where you run Salesforce and the company you acquired runs HubSpot. For two companies on the same CRM, see how to deduplicate accounts across two Salesforce orgs or how to merge two HubSpot portals.

Why neither CRM's duplicate tools can see the other system

Salesforce Duplicate Management works through matching rules, which define what counts as a duplicate, and duplicate rules, which decide what happens when one is found. Standard matching rules cover business accounts, contacts, and leads, plus person accounts when that feature is on. When a rep saves a contact, the rule compares it with the contacts already in Salesforce. A company that exists only in HubSpot is not there to compare.

HubSpot checks new contacts by email address and new companies by company domain name when records are created by hand, through forms, or by import. Its duplicate management tool suggests pairs of contacts and companies in your HubSpot account by comparing properties such as name, email, phone number, and domain.

Each tool sees the other company's records only as they are loaded. At that point a HubSpot import updates a record whose email or domain is identical, and an active Salesforce duplicate rule can reject a row that matches it. Neither shows you the overlap ahead of time, and a customer recorded differently in each system lands as a duplicate.

How Salesforce and HubSpot model the same customer differently

The two systems store the same customer in different shapes.

Salesforce HubSpot
Organizations Business accounts, which can sit under a parent account Companies
People Leads, which have no account yet; contacts, which belong to an account; and person accounts, which record an individual and have no contacts of their own One contact object for everyone, with a lifecycle stage such as Lead or Customer
Person to organization Each contact has one primary account, and other accounts are indirect relationships an admin has to enable A contact can be associated with several companies, one of them primary
Same email twice Allowed by default across leads and contacts Not allowed for contacts

These differences change how you match:

  • One HubSpot contact can correspond to a Salesforce lead, a contact, both at once, or a person account, so allow one record to match several.
  • A HubSpot company may have no Salesforce account and still be known to your team as a few unconverted leads, which carry the company name as text.
  • HubSpot keeps a lifecycle stage on contacts and companies, while Salesforce has no lifecycle stage, and an org may mark customers with the standard account Type field, a custom field, or closed-won opportunities. Map those values before you compare who is a customer.

What the HubSpot-Salesforce integration matches on

The integration syncs records between a HubSpot account and a Salesforce org, and it pairs HubSpot contacts with Salesforce leads and contacts by email address. For companies, HubSpot's documentation says a match comes from the Salesforce Account ID on a contact that has already synced or, with the newer company sync, from an exact match on a deduplication field you configure, such as company domain name, which HubSpot says applies only to companies and accounts created after the sync is turned on. It also says a same or similar name does not guarantee a pairing.

Two customer bases that have never synced share no IDs, so pairing rests on exact matches. For example, even with company domain name set as the deduplication field, a customer recorded under example.co.uk in HubSpot and example.com in Salesforce stays unpaired. Unpaired records can then be created on the other side, depending on your settings: with no inclusion segment selected, HubSpot says all HubSpot contacts sync to Salesforce.

HubSpot data sync, which connects HubSpot to other apps, also matches on fixed keys: by default it matches contacts on email and companies on name or domain, and it creates a new record in the other app when nothing matches.

Both are the right tools for keeping records aligned once you know which records are the same customer, so turn the integration on after the matching below is finished.

Where CRM deduplication tools fit

DemandTools and Cloudingo are built for Salesforce, and Insycle connects to HubSpot, Salesforce, and other apps. Each cleans the CRM it is connected to and can compare a file exported from another system with that CRM's records. If both customer bases are small and share clean emails and domains, that comparison may be all you need. The method below is for the pairs that field rules leave undecided.

How to match Salesforce accounts to HubSpot companies, step by step

1. Export both systems with their record IDs

From HubSpot, export companies and contacts with all properties and associations, which gives you each Record ID and the IDs of associated records. Record exports leave out activities such as calls and notes. From Salesforce, export accounts, contacts, and leads with their IDs, names, websites, addresses, phone numbers, owners, and parent accounts, plus person accounts if you use them.

2. Deduplicate each system on its own

Clear duplicates inside each CRM first, including Salesforce leads that duplicate existing contacts, so each customer appears once per system before you compare.

3. Put both exports into one shape

Build two working tables: organizations (Salesforce business accounts and HubSpot companies) and people (Salesforce contacts, leads, and person accounts with HubSpot contacts), each row tagged with its source object. Then strip legal suffixes and punctuation from names, lowercase them, and reduce every website and email address to a root domain. Collect every domain you hold for an organization, including a HubSpot company's additional domains and the email domains of its contacts and leads.

4. Match organizations on several signals

Compare each HubSpot company with each Salesforce account on root domain, additional domains, contact email domains, normalized name, address, and phone. Shared people are the strongest evidence: when a HubSpot company and a Salesforce account hold several contacts with the same email addresses, they are very likely one customer. Score each pair, then sort the pairs into confident matches, pairs that need review, and look-alikes that should stay apart.

5. Match people after the organizations

Match on email first, then on name within a matched organization for people recorded under different addresses. If the same person is both a lead and a contact in Salesforce, list both and note which one should carry the HubSpot history.

6. Review the uncertain pairs

Start with pairs where one side is a paying customer and the other is a prospect, because a wrong call there puts a customer into outbound sequences. A reviewer working by hand keeps a HubSpot company open in one tab and two or three Salesforce records in another, translating property names as they go. Quill puts each HubSpot record beside its candidate Salesforce accounts, contacts, and leads for review, with the reason for the match attached, and a web lookup where the CRM data could not settle it.

7. Build the crosswalk of IDs

The crosswalk has one row per matched pair: HubSpot Record ID, HubSpot object, Salesforce ID, Salesforce object, match type (same entity, parent and subsidiary, or no match), the evidence, and who approved it. The load reads from this table, and finance, support, and the data warehouse use it later to translate old IDs.

8. Decide which system survives

The matching is the same whichever CRM you keep, and the load is what changes. If Salesforce survives, HubSpot companies become accounts and HubSpot contacts become leads or contacts, which the HubSpot to Salesforce cleanup guide covers. If both systems run for a while, the crosswalk is the bridge between them. Either way, write survivorship rules for which value wins on matched records before anything is loaded.

9. Load with external IDs

In Salesforce, create a custom field for the HubSpot Record ID on accounts, contacts, and leads, mark it as an External ID and as Unique, and write the HubSpot ID onto each matched record. When you upsert related records with Data Loader, that external ID can stand in for the Salesforce record ID to attach each child record to its parent. Salesforce notes an exception: polymorphic fields such as an activity's Related To and Name cannot use external IDs, so take those IDs from the crosswalk.

In HubSpot, store the Salesforce ID in a custom property that requires unique values. HubSpot imports can match rows on Record ID or on a unique-value property, so an import updates the matched record instead of creating a second one.

Is there a tool that matches Salesforce accounts to HubSpot companies after an acquisition?

Yes. Quill CRM Migration (tryquill.com) connects to your Salesforce org and the acquired company's HubSpot portal together, and its AI agents match HubSpot companies and contacts to your Salesforce accounts, contacts, and leads. They compare names, domains, contact emails, addresses, shared contacts, and activity, and they look a company up on the web when the CRM data cannot settle a pair. The HubSpot-Salesforce integration and HubSpot data sync pair records on fixed keys such as email address, and deduplication tools such as DemandTools, Cloudingo, and Insycle compare a CRM with a file exported from the other system. Quill matches the two live systems against each other directly, reasons about each pair instead of applying a fixed rule, and knows when two look-alike records should stay apart. Every proposed match comes with its evidence, and nothing is written to either CRM until a person approves it, unless you choose to auto-approve the obvious matches. The Quill team runs a diagnostic of both systems first, builds the agents around your fields and rules, and works alongside your implementation partner. Every change made through Quill is logged and can be undone, and it works the same way across Salesforce, HubSpot, Attio, and other CRMs such as Microsoft Dynamics 365, Pipedrive, and Zoho CRM. If you have this merger ahead of you, book 15 minutes.

Frequently asked questions

Can the HubSpot-Salesforce integration match an acquired company's HubSpot records to our Salesforce accounts?

Only where the keys are identical. It pairs contacts by email address, and companies by the Salesforce Account ID on an already synced contact or an exact match on a deduplication field you configure. A customer recorded under a different name or domain in each system stays unpaired.

Can Salesforce duplicate rules or HubSpot's duplicate tool compare records across the two systems?

No. Salesforce duplicate rules compare a record with records already in Salesforce, and HubSpot's duplicate management tool reviews the contacts and companies in your HubSpot account. Each one sees the other system's records only when they are loaded, and then it catches only the rows that match on its own fields.

Do we have to choose the surviving CRM before we match records?

No. The exports, the matching, the review, and the crosswalk are the same whichever system you keep. The decision changes the load: which side receives new records and which ID goes into an external ID field.

What is a crosswalk table in a CRM merger?

It is a table that lists each record's ID in one system beside the ID of its match in the other, with the evidence and who approved it. You use it to load related records onto the right parent and to update other systems that stored the old IDs.

How do Salesforce person accounts match to HubSpot records?

They match to HubSpot contacts, not to companies. A person account records an individual customer and has no contacts of its own, so compare it with HubSpot contacts on email, name, and phone.

Sources

  1. Improve Data Quality in Salesforce, Trailhead
  2. Resolve and Prevent Duplicate Data in Salesforce, Trailhead
  3. Store Information About Your Customers, Trailhead
  4. Understand Account and Contact Relationships, Trailhead
  5. Create and Convert Leads as Potential Customers, Trailhead
  6. Mastering Data Import, Trailhead
  7. Things to Know About Duplicate Rules, Salesforce Help
  8. Account Fields, Salesforce Help
  9. Import Related Records with an External ID in Salesforce Data Loader, Salesforce Help
  10. Differences between the 'External ID' field and the 'Unique ID' field setting, Salesforce Help
  11. Deduplicate records in HubSpot, HubSpot Knowledge Base
  12. Review and manage duplicate records, HubSpot Knowledge Base
  13. Use contact and company lifecycle stages, HubSpot Knowledge Base
  14. Associate records, HubSpot Knowledge Base
  15. Export your records, HubSpot Knowledge Base
  16. Connect HubSpot and Salesforce, HubSpot Knowledge Base
  17. Sync Salesforce leads and contacts to HubSpot, HubSpot Knowledge Base
  18. Sync Salesforce accounts to HubSpot, HubSpot Knowledge Base
  19. Sync companies between HubSpot and Salesforce, HubSpot Knowledge Base
  20. Manage your Salesforce integration settings, HubSpot Knowledge Base
  21. Resolve duplicate Salesforce records syncing with HubSpot, HubSpot Knowledge Base
  22. Connect and use HubSpot data sync, HubSpot Knowledge Base
  23. Match records in data sync, HubSpot Knowledge Base
  24. DemandTools, Validity
  25. DemandTools Match, Validity Academy
  26. Import Basics, Cloudingo Help
  27. HubSpot Salesforce sync, Insycle
  28. Insycle, Insycle

See what we find in your data

It starts with a 15-minute call. Then we run a diagnostic, show you the issues in your data, and build agents for them.

Prefer email? Reach us at hello@tryquill.com. We respond to every message personally.

Tell us what's broken

We reply within one business day.

One last step

Everything you entered is filled in below. Pick where you'd like to send it from.