HubSpot and Salesforce, running together
Marketing runs HubSpot. Sales runs Salesforce. Both are staying. HubSpot Salesforce integration keeps them consistent so neither team has to change how they work and neither has to trust the other’s export.
HubSpot ships a native connector for this, which makes it one of the more mature integrations we deal with. The work is not making it connect. It is deciding, field by field, which system wins.
What syncs between HubSpot and Salesforce
The native connector covers the standard objects and rather more besides.
- Contacts map to Salesforce leads or contacts, with the choice of which being one of the first decisions.
- Companies map to Salesforce accounts.
- Deals map to Salesforce opportunities, including stage and amount.
- Activity history flows across, so marketing engagement is visible against the sales record.
- Custom objects can sync two-way, on the tiers that support it.
- Marketing engagement such as email opens, form submissions and page views lands against the contact.
The connector is available on HubSpot Professional and Enterprise. There are real limits worth knowing early: 500 field mappings per object, and up to ten custom objects per account.
Which way the data flows between HubSpot and Salesforce
This is the most two-way integration in the set, which is why it needs designing carefully. Every field can be set independently.
⇆ Two way→ Out of Salesforce← Into Salesforce
HubSpot lets you set each field mapping to Two-way, Always use Salesforce, or Always use HubSpot. That flexibility is the feature and the trap. Left on defaults across hundreds of fields, you get update loops and quiet overwrites nobody notices for weeks.
What a HubSpot integration gives you
Two systems that agree, and a marketing function that can prove its contribution.
- Sales sees what marketing did. The rep opens the record and can see which campaigns the contact engaged with, rather than asking.
- Marketing sees what closed. Closed-won data flowing back is what turns campaign reporting into revenue attribution.
- Lead handover stops being a spreadsheet. Qualified leads arrive in Salesforce with their history attached.
- One consent position. Opt-outs propagate rather than living in whichever system happened to record them.
- No duplicate maintenance. Nobody updates the same company in two places, and reporting stops diverging.
How we connect HubSpot to Salesforce
The question to settle before you build the HubSpot integration
Which system owns the contact, and at what point does ownership change hands?
A contact usually starts life in HubSpot as a marketing record and becomes a Salesforce record when sales picks it up. The integration has to know where that line sits, because it decides which system wins when both have been edited.
Get it right and the handover is invisible. Get it wrong and you have marketing overwriting a rep’s correction every time a campaign runs, which is the single most common way this integration loses people’s trust.
The same question applies to lifecycle stage, because both platforms automate off it. Two systems both setting lifecycle stage will fight, quietly and continuously.
The lead that marketing swore was qualified
Marketing hands over a lead scored 92. Sales calls it, discovers the contact left the business in January, and stops trusting the score. Nobody says so out loud, and the two teams quietly go back to working separately.
That failure is almost always a sync failure rather than a scoring failure. The contact updated their details in Salesforce, HubSpot never heard, and the score was calculated against a person who no longer exists.
A properly configured two-way sync fixes the specific problem and a broader one. Sales sees which campaigns the contact actually engaged with before dialling, so the call opens with something relevant. Marketing sees closed-won flowing back, so campaign reporting becomes revenue attribution rather than an activity report.
The configuration decision that matters is which system owns lifecycle stage. Both platforms automate off it. Allowing both to set it produces a quiet, continuous conflict. The first symptom is usually a rep reporting that a contact keeps reverting to a stage they corrected last week.
This is the integration where the technology is easiest and the internal negotiation is hardest. Budget accordingly.
Connect HubSpot to Salesforce
The connector is the easy part. Agreeing who owns what, field by field, is the work, and it is what makes the difference between an integration people trust and one they route around.
Whatever you run alongside it, if it has an API it can be connected, so get in touch and we will tell you what is realistic.
HubSpot Salesforce integration: frequently asked questions
Yes. The native connector syncs contacts, companies, deals, activity history and custom objects in real time, and each field mapping can be set independently to two-way, always use Salesforce, or always use HubSpot. That per-field control is the point of it, and configuring it deliberately rather than accepting defaults is what makes the difference.
The main ones are 500 field mappings per object and up to ten custom objects per HubSpot account. Custom object sync is not available on every subscription tier, and the connector requires HubSpot Professional or Enterprise. Where you need more than that, middleware handles it, and we will tell you if you are close to a limit before you commit.
Not if matching is designed first. Duplicates come from syncing on email address alone, or from starting before anyone agreed what makes two records the same. We define matching rules, run them against your real data and show you the collisions before anything is switched on.
Yes, and that is the most common reason to do this. Neither team has to move. HubSpot keeps the marketing automation and Salesforce stays the system of record, with a defined boundary between them so the same contact is not maintained twice.
A standard connection covering contacts, companies and deals is usually days rather than weeks. What extends it is the number of custom fields, the state of the existing data, and how much disagreement there is about ownership. The data clean-up is almost always the longest part.