The seven things worth syncing between Zendesk and HubSpot: contact and company identity, CRM context on tickets, conversation threads, AI summaries, deal flags from tickets, custom object data, and your historical backlog. That's the whole playbook — the rest of this post is the "so what" for each.
Plenty of teams get as far as "we should integrate these" and stall, because the idea arrives without a spec. Having built a purpose-built Zendesk ↔ HubSpot sync, we can tell you exactly which syncs earn their keep. Here they are, in the order most teams should turn them on.
What's the foundation every sync needs?
Identity first — nothing else works until both systems agree on who the customer is.
1. Contact and company identity
Sync contacts and organizations bidirectionally so a person in Zendesk and a person in HubSpot are the same record, matched and kept current in real time. So what: every other sync on this list rides on this rail. Get it wrong and you're syncing tickets onto duplicate ghosts. Get it right and updates made on either side — a new phone number, a corrected company name — land on the other without anyone retyping anything.
2. Lifecycle and deal context onto tickets
Pull HubSpot lifecycle stage, deal stage, and deal value into Zendesk as context tags on the ticket. So what: your agents stop treating a six-figure renewal account like a free-trial signup — not because they don't care, but because until now, Zendesk literally couldn't tell them the difference. Context tags turn "a ticket" into "a ticket from a customer in active renewal," which changes tone, priority, and escalation instincts instantly.
How should conversations flow into the CRM?
Cleanly, or not at all — a noisy sync is worse than none, because it trains sales to ignore the timeline.
3. Ticket conversations as one clean thread
Log each ticket to the HubSpot record as a single, tidy conversation thread — not one CRM activity per reply. So what: a rep prepping for a call reads the support situation in thirty seconds instead of scrolling past forty fragments. The renewal-call-during-an-outage disaster becomes structurally impossible, because the outage is right there on the record.
4. AI summaries, on demand
Let agents share an AI-generated summary of a ticket to HubSpot when a situation deserves executive visibility. So what: some tickets are sagas. When the account manager needs the plot, an agent-initiated summary delivers "customer hit a billing bug, was frustrated, resolved with a credit, watch for sentiment at renewal" — instead of homework. Agent-initiated is the key word: humans decide what's worth elevating.
Which syncs connect support to revenue?
These two — they're where the integration stops being convenient and starts being profitable.
5. Deal flags from tickets
Optionally create or flag a HubSpot deal from a Zendesk ticket when support uncovers revenue — an expansion question, a plan-limit complaint, an "is there an API for this?" So what: support hears buying intent all day and has nowhere to put it. An opt-in ticket-to-deal path gives those signals a pipeline instead of a shrug. Opt-in matters: nobody wants a CRM full of auto-generated deal confetti.
6. Custom object data
Sync the weird stuff — subscriptions, devices, projects, whatever custom objects your business actually runs on. So what: for most companies the real customer context isn't in standard fields. An agent who can see which subscription tier or which installed asset the ticket concerns skips three rounds of "can you confirm your plan?" Rigid connectors can't touch this; field mapping with transforms makes it routine.
What about everything that happened before the integration?
Sync it too — history is where the patterns live.
7. Backfill of historical tickets
Bulk-backfill past Zendesk tickets onto the matching HubSpot records so the integration starts with memory, not amnesia. So what: without backfill, every record looks like the customer showed up yesterday. With it, the rep sees two years of context on day one — and the "repeat issue" and "rising frequency" churn signals are actually detectable, because there's a baseline to rise from.
Do you need all seven on day one?
No — but you need one, three, and two, in roughly that order. Identity is the rail, clean threads are the payoff sales notices first, and context tags are the payoff support notices first. Deal flags, summaries, custom objects, and backfill layer on as the team trusts the data.
The good news: with a real-time bidirectional sync and visual field mapping, "layering on" is configuration, not a new project. Everything above ships in our integration — OAuth-connected, signed webhooks, and designed to respect read-only fields so it never bulldozes data you've told it not to touch.
Want the playbook mapped to your actual pipelines and ticket fields? Talk to us — scoping this is genuinely our favorite conversation.
Frequently Asked Questions
What should I sync between Zendesk and HubSpot?
Start with contact and company identity, then ticket conversations logged to CRM records as clean threads, and HubSpot lifecycle and deal context surfaced on tickets. Mature setups add deal creation from tickets, on-demand AI summaries, custom object data, and a backfill of historical tickets.
Can Zendesk tickets create deals in HubSpot?
Yes, with the right integration — and it should be opt-in, not automatic. When a support conversation reveals expansion interest, an agent or workflow can flag it into a HubSpot deal. Automatic ticket-to-deal creation tends to flood the pipeline, so human judgment on which tickets become deals is the pattern that works.
Does a Zendesk-HubSpot integration sync historical tickets?
Purpose-built integrations can. A bulk backfill attaches past ticket history to the matching HubSpot contacts and companies, so records show the full relationship from day one. Basic connectors and middleware usually only sync from install-date forward, leaving every record looking brand new.