HubSpot to BigQuery connector pricing: what Google's free Preview transfer, HubSpot, Fivetran and Airbyte really cost

How much does the BigQuery HubSpot connector cost?

Zero dollars for the transfer, today. Google's setup page says it plainly: "There is no cost to transfer HubSpot data into BigQuery while this feature is in Preview." The BigQuery pricing page adds the rule behind it. Connectors in Preview "will not be charged", and billing starts once a connector is generally available. HubSpot is listed there as a Preview connector, beside Stripe, PayPal, Klaviyo and SQL Server.

Free comes with terms. Preview features run under Google's Pre-GA Offerings Terms, which describe them as available "as is" with limited support, and support for this connector goes through a preview mailbox rather than your normal support plan. For a dashboard the sales team glances at, that is a fine trade. For board metrics, keep a fallback you have actually tested. Every other documented limit of the connector is on HubSpot to BigQuery integration tools, next to HubSpot's own integration.

What will the Google HubSpot connector cost after Preview?

Google has not published a rate, so any number is a forecast. What is published is the model. The pricing page puts HubSpot among the paid connectors, which are consumption-based and measured in slot-hours. The generally available ones, Salesforce among them, cost $0.06 per slot-hour on demand in us-central1, and Google gives a planning guideline of up to 20 slot-hours per hour a transfer runs. That is about $1.20 per run hour at the ceiling. We worked through that meter for the other CRM in Salesforce to BigQuery connector pricing.

The HubSpot connector has one property the Salesforce one does not, and it changes the forecast. Google's page states that "with every transfer run, the HubSpot connector transfers all available data from HubSpot into BigQuery". No incremental mode or watermark is documented. On a run-time meter, a full reload means each run costs roughly what the whole portal costs to read, every time. If HubSpot is priced on the same basis at GA, a daily schedule on a mid-size portal stays cheap, and an hourly one on a large portal is 24 times the run hours for the same data. The practical step today: export your run durations from the transfer history each month, so you can price your own workload the day Google publishes a rate.

How many HubSpot API calls does a BigQuery sync use?

This is the cost that shows up in HubSpot rather than on a Google invoice. Private apps in a portal share a daily API allocation: 250,000 calls on Free and Starter, 625,000 on Professional and 1,000,000 on Enterprise, per HubSpot's usage guidelines. HubSpot's changelog once gave 650,000 for Professional, and the guidelines table says 625,000, so we use the lower figure. Google does not document which endpoints its connector calls. The table below assumes 100 records per call, the page size of HubSpot's CRM list and batch endpoints, for a portal holding 600,000 records across contacts, companies and deals. The call counts are our arithmetic, not a measurement.

A quarter of the daily allowance sounds survivable until you list what else draws on it: the enrichment tool, the billing sync, the support desk integration and whatever the marketing team installed last quarter. When the pool runs dry they start receiving 429 errors, and none of them will say the warehouse caused it. The last row is the comparison that matters. An incremental sync reads changed records, so its call count follows your activity rather than your database size. How the burst and daily limits interact is covered in HubSpot API rate limits.

The private app token is worth one more line in the budget, because it carries a deadline and a security review. Google's connector requires more than 30 scopes, including scope names that describe importing and exporting CRM data, and HubSpot stops creating legacy private apps for existing accounts on October 26, 2026, with a migration to its Projects model due by September 2027. Whoever signs off the pipeline will want the token's owner, scopes and rotation schedule written down, and teams that already track vendor access for SOC 2 can record it as one more control in their compliance management software rather than a note in someone's inbox.

Is Fivetran or Airbyte cheaper for HubSpot to BigQuery?

They count different things, so the same portal can be cheapest on either. Fivetran counts rows that changed during the month, once each however often they changed. Airbyte counts rows synced, at 6 credits per million for API sources such as HubSpot, and on a full refresh it counts every row on every run. Google counts run time and reloads everything. HubSpot's own integration counts nothing extra, because it is part of a subscription tier. The table puts the six counting rules side by side.

Put numbers on it with the same example portal: 600,000 records, of which about 20,000 change on a normal day. Airbyte Cloud's Standard plan is $20 a month with 5 credits included and extra credits at $5 each. Synced incrementally, 600,000 changed rows a month is 3.6 credits and fits inside the $20 plan. Switch the same portal to a daily full refresh and it becomes 18 million rows a month, which is 108 credits, or roughly $535 on the Standard plan by our arithmetic. Airbyte's current plans, including what happens when credits run out, are on Airbyte pricing.

Fivetran does not publish a per-row rate you can plug in, so we will not invent one. Its monthly active row model forgives churn: a deal edited thirty times in a month counts once. It punishes the first month, when every historical contact and deal is active because it was just loaded, so price that month on its own. How the meter behaves is covered in Fivetran pricing.

Is HubSpot's own BigQuery integration free?

It is included, on one tier. HubSpot's Google BigQuery integration is a beta available only with Data Hub Enterprise, and HubSpot says credits apply when data is prepared in Data Studio first. If you already pay for that tier, the integration costs nothing extra and exports objects, custom objects, associations and events. If you do not, buying the tier to get the integration is a much larger decision than choosing a connector. HubSpot's pricing page lists the tier's price, and we will not quote it second-hand.

Two limits change the value. The fastest schedule during beta is every 6 hours, and HubSpot states that sensitive data cannot be synced to BigQuery during the beta. It also lands data as parquet files in a Cloud Storage bucket behind BigQuery external tables, so you pay Cloud Storage for the files and query external tables that are slower than native storage. Heavy dashboards usually end up copying them into native tables on a schedule, which is one more job to own.

What else do I pay for besides the connector?

BigQuery storage for the landed tables and compute for the queries that read them. On-demand queries cost $6.25 per TiB scanned, with the first TiB each month free, per Google's pricing page. A full reload each run also rewrites the tables, and storage only drops to the cheaper long-term rate for tables not modified in 90 days, so a table reloaded daily always bills at the active rate. For a CRM portal these numbers are usually small next to the people cost of reconciling wrong totals, which is why the type decisions matter more than the storage line.

When is a flat price cheaper than a free connector?

When the free connector's real costs are the API allowance, a Preview support path and a future rate nobody can forecast. Adapters charges a flat tier by records synced: Starter at $49 for 10,000 records a month with hourly syncs, Growth at $149 for 100,000 with 5 minute syncs, and Scale at $399 for 1 million with 1 minute syncs. There is no overage fee, and reading records that have not changed costs nothing, so after the backfill the example portal's 20,000 daily changes land on Scale.

Be clear about where that loses. On price alone, Google's connector at $0 beats every paid option today, and Airbyte's $20 plan beats us for a quiet portal synced incrementally. What a flat plan buys is a number that does not move when Google leaves Preview, a sync that reads only changes, and control over what lands: amount as NUMERIC and closedate as DATE, rather than a close date stored as a midnight UTC timestamp that slips to the prior evening in a US report. If that matters more than the last few hundred dollars, the HubSpot to BigQuery connector is set up in minutes. Every other meter in the market is on data integration pricing.

How do I keep the HubSpot to BigQuery bill down?

Match the schedule to the decision the data feeds. Pipeline reviews are weekly and forecasts daily, so a daily full reload covers most revenue reporting, and it costs a small fraction of the API allowance. Put slow objects such as owners, pipelines and products in their own daily transfer, and keep the faster schedule for deals and contacts only. Never run an hourly full reload on a large portal just because the transfer is free today, because the same schedule is what you will be billed for at GA. The same reasoning for the other warehouse is on HubSpot to Snowflake integration tools, and every other source into BigQuery is on BigQuery ETL tools.