What Goes Wrong When You Run HubSpot and Salesforce at the Same Time

The setup seems logical. Marketing lives in HubSpot because the automation, landing pages, and email tools are better. Sales lives in Salesforce because that is what the CRM team already knows. HubSpot's native Salesforce integration syncs the data between them. Everyone is happy. Until they are not.

Running HubSpot and Salesforce in parallel is a legitimate architecture for some B2B companies. But it creates a specific set of integration problems that are predictable, recurring, and difficult to resolve without deliberate effort. This post covers the five most common problems and what to do about each one.

Why Companies End Up Running Both

The most common path to a dual-system setup is not a deliberate decision. It happens incrementally. The marketing team adopts HubSpot for email and automation while sales is already on Salesforce. Someone connects the integration. Both teams continue working in their respective systems. The integration mostly works, and nobody pushes to consolidate because it would require someone to give something up.

The integration continues working mostly well until the company needs something more specific from it, at which point the underlying problems surface.

Problem 1: Duplicate Records

The most common and most damaging problem with the HubSpot-Salesforce sync is duplicate contact records. It happens when a contact exists in Salesforce but was created differently in HubSpot, or when the inclusion list that controls which contacts sync is set too broadly.

The default sync behavior tries to match on email address. But if someone is in Salesforce as john.smith@company.com and in HubSpot as j.smith@company.com because they filled out a form with a different format, the sync creates a second record instead of merging. Now the same person exists twice, with different activity histories in each system, and sales is calling a contact that marketing already followed up with three times this week.

The fix: Implement HubSpot's deduplication workflow before enabling the sync. Set the sync inclusion list to a specific Salesforce report rather than all contacts. Run a deduplication pass on both systems quarterly.

Problem 2: Lifecycle Stage Conflicts

HubSpot has a lifecycle stage field (Subscriber, Lead, MQL, SQL, Opportunity, Customer). Salesforce has its own lead and contact status fields, plus opportunity stages. These fields do not map natively to each other, and the sync does not know which system should win when they conflict.

The result: a contact gets marked as Customer in HubSpot when they close in Salesforce, but the sync runs and overwrites the HubSpot lifecycle stage back to SQL because that was the last value that synced from Salesforce. Now HubSpot is sending that contact prospecting emails because the system thinks they are still a prospect.

The fix: Define a sync rule that makes HubSpot the source of truth for lifecycle stage and Salesforce the source of truth for opportunity stage. These are different concepts and should not compete. Document the field mapping explicitly and enforce it through sync rules rather than hoping both systems agree.

Problem 3: Sync Lag Causing Automation Failures

The standard HubSpot-Salesforce sync runs every 10 to 15 minutes. For most data, this is fine. For automation that needs to react to a specific trigger, it is not.

The failure scenario: a Salesforce deal closes and moves to Closed Won. A HubSpot workflow is set to trigger when the lifecycle stage becomes Customer and enroll the contact in an onboarding email sequence. The workflow fires based on HubSpot data, but the deal close has not yet synced from Salesforce to HubSpot. So the workflow either fires 15 minutes late or, if something else changed in the contact record in the meantime, misses the trigger window entirely.

The fix: For time-sensitive automation, trigger from HubSpot data that changes in HubSpot, not from data that arrives via sync. If you need to trigger on a Salesforce event, use a Salesforce workflow to push the event via webhook rather than waiting for the sync cycle.

Problem 4: Attribution Falls Apart

Marketing attribution requires knowing which touchpoints contributed to a closed deal. In a single-CRM setup, that data lives in one place. In a dual setup, the marketing touchpoints are in HubSpot and the deal is in Salesforce. The native integration syncs some engagement data back to Salesforce, but the attribution model in HubSpot cannot see the full Salesforce opportunity history, and the attribution model in Salesforce cannot see the full HubSpot engagement history.

The result is that neither system gives you accurate first-touch, last-touch, or multi-touch attribution. Marketing cannot defend its contribution to revenue because the data is split. Sales cannot understand which marketing programs are generating the deals that close because the insight requires joining two datasets that do not join cleanly.

The fix: Define which system owns attribution reporting and build your attribution model exclusively in that system. Most B2B teams choose HubSpot for marketing attribution and Salesforce for revenue reporting, which means accepting that the two numbers will never match exactly. The alternative is a third-party attribution tool like Dreamdata or Rockerbox that pulls from both systems, which adds cost and complexity.

Problem 5: Contact Owner Conflicts

Both HubSpot and Salesforce have a Contact Owner field. When a lead is assigned in HubSpot, the owner may or may not match the Salesforce owner. When the sync runs, the field conflict can silently overwrite one system's owner assignment with the other's.

This creates a real operational problem: a HubSpot sequence is running from rep A's email address, but the Salesforce record shows rep B as the owner. Rep B's manager asks why rep B has not followed up on this contact. Rep A did not know the record was also in Salesforce. The contact gets called by both reps in the same week.

The fix: Designate one system as the source of truth for owner assignment and set the sync to only flow in one direction for that field. If Salesforce owns territory and rep assignments, HubSpot should receive owner from Salesforce, not overwrite it.

When to Stop Running Both

The integration problems above are solvable. But solving them requires ongoing governance work. The question is whether that work is worth the benefit of maintaining both systems.

The case for consolidating to one platform becomes clear when:

• Your RevOps or admin team spends more than a few hours per week managing sync issues
• Sales and marketing leadership cannot agree on which system's data is correct
• New employee onboarding requires training on two separate systems
• A pipeline or attribution question cannot be answered without pulling data from both systems and reconciling it manually

The integration between HubSpot and Salesforce is one of the most robust CRM integrations on the market. It works well when it is configured deliberately and governed actively. It creates compound data problems when it is connected and left to run without oversight. Most of the companies we work with that are unhappy with the dual setup are not suffering from a bad integration. They are suffering from an integration that was connected and never properly configured.