QuickBooks to BigQuery integration and connector tools compared, with the limits that decide the choice
BigQuery has no native QuickBooks connector, so every route is a tool you add. Since November 2025 Intuit also meters the read calls a warehouse sync is made of, so the tool that reloads everything every hour now costs more than the one that reads only changes. Read from Intuit and Google Cloud documentation on 17 November 2026.
The live demo needs no card, and Starter is $49 a month.
Field mapping auto-plugged · tap a port to rewire
Plug a source port into
Transform on this cable
JSON in
JSON out
5 sample records ready
Vendor documentation read 17 November 2026
Which QuickBooks to BigQuery connector should you use?
Use a scheduled connector that backfills history with paged queries, then reads only changes through QuickBooks change data capture, and lands amounts as exact NUMERIC columns. That pattern stays inside the 10 connection limit per company, spends the fewest metered CorePlus reads, and produces tables you can tie to your profit and loss. CSV exports suit a one-off analysis, and trigger based automation suits an alert table, but neither gives finance a model that reconciles.
For QuickBooks on its own, including every API limit that governs an integration, see QuickBooks integration tools. For every other source into the same warehouse, see BigQuery ETL tools, and for an ERP into BigQuery, NetSuite to BigQuery integration tools.
What the Intuit and BigQuery documentation say, and what a buyer assumes
We read Intuit's API limits and App Partner Program pages beside Google's transfer and quota pages on 17 November 2026 and wrote down what each line costs an accounting team. The last column is what we would do.
| What is at stake | What the docs say | Why it matters | What we would do |
|---|---|---|---|
| BigQuery side | The BigQuery Data Transfer Service lists Stripe, PayPal and Shopify in Preview, and Salesforce, HubSpot and ServiceNow. It lists no QuickBooks or Intuit source. | Your payments can arrive through Google while the ledger they settle into cannot. Every QuickBooks route is a tool you add. | Budget a tool, not a toggle |
| Intuit metering | Core calls (data in) are unmetered. CorePlus calls (data out: reads, queries, reports) are metered since 3 November 2025. Only successful 2xx production calls count. | A warehouse sync is read traffic, so it is exactly the class Intuit meters. A tool that reloads every table every hour spends credits it does not need to. | Ask how many reads a run costs |
| Builder ceiling | The free Builder tier includes 500,000 CorePlus credits a month per workspace, and calls above that are blocked until the 1st. | If a small vendor runs every customer through one Builder workspace, the sync can stop mid-month for reasons you never see. | Ask which partner tier the vendor holds |
| Concurrency | 500 requests per minute and a maximum of 10 concurrent connections per company, scored on the app plus company combination. | Ten connections is the ceiling that bites. Scaling a worker pool out produces HTTP 429 ThrottleExceeded at low volume, and retries make it worse. | Cap parallelism at a few connections |
| Query pages | Queries return 100 records by default and at most 1,000 with MAXRESULTS, paged by STARTPOSITION. | Nothing warns you that more records exist. An unpaged load reports success and keeps the first page only. | Page every query, verify counts |
| BigQuery table limit | A standard table accepts 1,500 modifications a day, and every load job counts, failed ones included. | A pipeline that loads once per page of a backfill can burn the daily budget on one table before lunch. | Stage a window, issue one load |
The metering row is the one most buyers have not heard of. Intuit classifies each call by endpoint and HTTP method, and its API classification for the App Partner Program puts reads, queries and reports in the metered class. The concurrency math behind the other rows is worked through on QuickBooks API rate limits.
QuickBooks to BigQuery connectors and integration tools compared
Billing units rather than price tags, because most vendors here meter or quote and any figure we printed would be stale within a quarter. Where a tool is the wrong fit, the last column says so, including for us.
| Tool | Approach | Best for | Billing unit | Watch for |
|---|---|---|---|---|
| Manual CSV export | Export QuickBooks reports to CSV and load them into BigQuery by hand | A one-off analysis or a single historical snapshot | No tool cost, paid in staff hours every month | Report exports are summaries, not line level, and nobody owns the refresh |
| Fivetran | Managed ELT with a QuickBooks Online connector writing tables into BigQuery | Teams with a data engineer and many sources already on the platform | Metered by monthly active rows, confirm on the vendor pricing page | Invoice-heavy books raise the bill, so model your busiest month |
| Airbyte | Open source or cloud ELT with a QuickBooks source connector | Engineering teams comfortable running and tuning pipelines | Self-hosted or usage-based cloud, confirm current units with the vendor | Self-hosting means you own OAuth refresh, retries and upgrades |
| Coupler.io | Scheduled QuickBooks report and entity exports into BigQuery | Analysts refreshing a handful of reports for a dashboard | Subscription by data flows and refresh frequency, confirm on the vendor page | Built around report extracts, so check line history and deletes |
| Hevo Data | No-code pipelines from QuickBooks Online into BigQuery | Analysts who want setup without code at modest volume | Metered by events, confirm tier limits with the vendor | Event billing punishes full reloads of large invoice tables |
| Workflow automation | Trigger based tools that insert one BigQuery row per new QuickBooks record | A single alert or log table, not a reporting model | Billed per task or operation, confirm on the vendor page | No backfill, no updates to old rows, no deletes, so totals drift |
| Adapters | Visual field mapping of QuickBooks entities onto typed BigQuery tables, paged backfill, change data capture after it | Finance teams that want NUMERIC money columns and a fixed monthly price | Flat monthly plans from $49, no per-row or per-event charge | Not built for QuickBooks Desktop files, which use a different API |
Vendor billing units change, so confirm each one on the vendor's own page before you model a month. Our plans are on pricing, and the cost of the whole category is worked through on data integration pricing. If what you actually need is the trigger based row, one record at a time into a sheet or a chat channel, a workflow automation product such as Adapter is the better fit, and a warehouse connector is overkill.
Six facts worth knowing before you choose
0
QuickBooks or Intuit sources in the BigQuery Data Transfer Service list on 17 November 2026.
docs.cloud.google.com
3 Nov 2025
The date Intuit began metering CorePlus calls, the read class every warehouse sync uses.
Intuit App Partner Program
500,000
CorePlus credits a month on the free Builder tier, after which calls are blocked until the 1st.
Intuit App Partner Program
10
Maximum concurrent connections per QuickBooks company, the limit syncs actually hit first.
Intuit, API call limits and throttles
100
Default records per query page, returned with no warning that more exist. MAXRESULTS goes to 1,000.
Intuit, Accounting API query
1,500
Modifications per BigQuery table per day. Each load job counts, failed ones too.
BigQuery quotas and limits
Six ways a QuickBooks to BigQuery pipeline is wrong while the log is green
Finance finds these at month end, not in monitoring. Each one exits zero.
| Failure | Why it is silent | The check |
|---|---|---|
| First page only | A query without STARTPOSITION paging returns 100 records and exits zero. The table looks populated. | Compare invoice counts by month against a QuickBooks report after the backfill. |
| Deleted and voided transactions | A sync that reads changes since a timestamp never removes what was deleted, so revenue drifts upward in BigQuery. | Ask the vendor which column flags a deleted record and test by voiding one invoice. |
| Amounts stored as FLOAT64 | Binary floating point cannot hold most cent values exactly. Large sums drift by cents against the P&L. | Sum one closed month in both systems. It must match to the penny. |
| Expired refresh token | QuickBooks OAuth refresh tokens rotate. A sync that does not store the new one fails quietly over a weekend. | Alert on the time since the last successful run, not just on errors. |
| Credits exhausted mid-month | Above the Builder allowance Intuit blocks CorePlus calls until the 1st. The pipeline stalls with nothing wrong on your side. | Watch the last loaded timestamp per table and ask the vendor about its tier. |
| Lost account hierarchy | Loading accounts without the parent reference flattens the chart, so nothing rolls up the way your controller expects. | Rebuild the P&L in BigQuery from the account table and compare subtotals. |
Six steps that decide whether this project works
Step 1
List the entities you report on
Invoices, payments, credit memos, bills, bill payments, journal entries, items, customers, vendors and the chart of accounts cover most finance models. Loading only invoices gives you a revenue chart nobody can reconcile.
Step 2
Declare money columns first
Create amounts as NUMERIC, document dates as DATE, and partition the large transaction tables by document date so a quarter-end query scans one quarter rather than the whole history.
Step 3
Backfill with paged queries
Pull history one entity and one date window at a time at 1,000 records per page, and tie each month to a QuickBooks report before moving on. Change data capture cannot do this step.
Step 4
Switch to change data capture
After the backfill, read only what changed since the last run. It is the cheapest pattern on CorePlus credits and it stays far inside 500 requests per minute.
Step 5
Keep concurrency low
Run a few connections, not a pool of twenty. The 10 connection ceiling is shared with your payroll, payments and expense apps on the same company.
Step 6
Reconcile after every close
Count and sum by month in both systems. A mismatch is a failed sync even if the log is green, because finance only finds it when a board number is wrong.
With Adapters the type decisions in step 2 happen when you map the field, so an invoice line amount lands as NUMERIC and a transaction date as DATE from the first run, and every run after the backfill reads change data capture only. The mapping for this pair is on the QuickBooks to BigQuery connector page, and the full cost of a QuickBooks warehouse, with the read count per schedule, is in QuickBooks data warehouse cost.
Five jobs where you should not pick us
A comparison that never says the competition wins is an advert. These are the cases where something else on this page is the right answer.
- Your books are in QuickBooks Desktop. It uses a different integration path, and this page and our connector cover QuickBooks Online.
- You already run Fivetran or Airbyte across thirty sources and QuickBooks is one more. Adding a second tool for one source rarely pays back.
- You want a finished BigQuery data model with dbt packages maintained for you. That is a managed ELT vendor with a modeling team.
- You need one report refreshed once a month for a single spreadsheet. A scheduled CSV export is cheaper than any tool here.
- You need real-time streaming of every edit. QuickBooks change data capture is polled, so every route here, ours included, runs on a schedule.
Four questions to ask any vendor on this list
Question 01
What exactly lands in BigQuery?
Ask for the destination schema of the invoice line table and the column type used for amounts. Typed NUMERIC columns and a JSON blob per invoice are different products under one name.
Question 02
How many CorePlus reads does one run cost?
Ask whether the tool reloads full tables or reads change data capture, and which Intuit partner tier it runs on. That decides whether your sync can be blocked late in the month.
Question 03
How are deletes and voids handled?
Ask for the behavior by entity and the column that marks a deleted record. A vendor that cannot answer has not handled it.
Question 04
What happens on a 429?
Ask how many concurrent connections the tool opens per company and whether you can cap it. Backoff with more workers is the wrong answer on QuickBooks.
Related QuickBooks and BigQuery guides
QuickBooks integration tools
Every QuickBooks Online limit and the wider vendor landscape.
QuickBooks data warehouse cost
Tool, metered Intuit reads and warehouse charges, worked out per schedule.
BigQuery ETL tools
Every other source into the same warehouse, and how each load is billed.
QuickBooks to BigQuery connector
Open the preset, review the field mapping and set the schedule.
QuickBooks API rate limits
Why the 10 connection cap makes a scaled-out sync slower.
QuickBooks to Snowflake
The same books into the other big cloud warehouse.
Questions buyers ask about QuickBooks to BigQuery integration
- How do I connect QuickBooks to BigQuery?
- BigQuery cannot read QuickBooks Online directly, so a tool has to call the QuickBooks API and load the rows. You can export CSV reports by hand, write a script against the Accounting API, or use a connector that maps entities onto BigQuery tables on a schedule. For anything finance reconciles monthly, a scheduled connector with typed columns is the only route that holds up.
- Does BigQuery have a native QuickBooks connector?
- No. The BigQuery Data Transfer Service source list read on 17 November 2026 includes Stripe, PayPal and Shopify in Preview, plus Salesforce, HubSpot and ServiceNow, but no QuickBooks or Intuit entry. So the payment side of your revenue can land through Google, and the ledger side always needs a vendor or a script.
- Does Intuit charge for QuickBooks API calls?
- For reads, yes. Under the Intuit App Partner Program, Core calls that write data in are unmetered, while CorePlus calls that read data out are metered since 3 November 2025. The free Builder tier includes 500,000 CorePlus credits a month and blocks calls above that. A warehouse sync is almost entirely reads, so it is CorePlus traffic.
- How much does a QuickBooks to BigQuery connector cost?
- Three costs stack. The connector subscription, any Intuit API fee your vendor passes on or absorbs for CorePlus reads, and BigQuery storage and query charges. Most pipeline vendors meter by rows, events or volume, which grows with your invoice count. Adapters is a flat $49, $149 or $399 a month by plan, with no per-row charge.
- How often can QuickBooks data sync to BigQuery?
- QuickBooks Online allows 500 requests per minute and 10 concurrent connections per company, shared with every other app connected to that company. BigQuery allows 1,500 table modifications a day per table. An hourly change data capture call is a few hundred reads a month per company, well inside both limits, and fast enough for finance reporting.
- Why are rows missing from my QuickBooks export in BigQuery?
- The most common cause is pagination. A QuickBooks query returns 100 records by default with no warning that more exist, and at most 1,000 per page with MAXRESULTS. A script that never pages with STARTPOSITION silently loads the first 100 invoices. Compare row counts per month against a QuickBooks report before you trust any load.
- Can change data capture backfill QuickBooks history into BigQuery?
- No. The changedatacapture endpoint returns up to 1,000 changed objects per call with no pagination, and it rejects a changedSince date older than Intuit's fixed retention window. It keeps a sync current after the first load. The history itself has to come from paged queries, one entity and one date window at a time.
- What is the best QuickBooks to BigQuery integration for an accounting team?
- The one that lands invoice and bill lines with exact NUMERIC amounts, keeps the chart of accounts hierarchy, handles deleted and voided transactions, and reads changes only rather than full reloads, which keeps CorePlus usage low. Run any trial on one closed month and tie the totals to your QuickBooks profit and loss to the penny.
For pushing warehouse results back into QuickBooks or your CRM, see reverse ETL tools, and for the whole market in one place, the best data integration tools.
Land QuickBooks in BigQuery with amounts that tie to your P&L
Map each entity once, declare exact decimals, backfill with paged queries, then read only what changed, with retries, alerts and per-record logs. Flat plans from $49 a month, no overage fees.
The live demo needs no card, and Starter is $49 a month.