HubSpot Data Hub (Formerly Operations Hub): Tiers, Catalog Changes, Programmable Automation and Setup
HubSpot Data Hub, formerly Operations Hub, explained tier by tier: data sync, formatting, datasets, programmable automation, permissions, credits and setup.
Paul Maxwell, PhD
AUTHOR
GET WEEKLY REVOPS INSIGHTS
No spam. Unsubscribe anytime.
HubSpot Data Hub is the name HubSpot now gives the product it sold as Operations Hub, and three expensive mistakes follow from reading it by name alone. A buyer who wants custom-coded workflow actions adds Data Hub Starter at $20 a seat, then finds that code and webhooks begin at Professional, an $800 monthly base. An administrator accepts the data quality tool's fix for a batch of capitalised names, and the sync that wrote them restores the capitals at each customer's next change in the source system, with nothing flagged. A team schedules an hourly dataset sync, and it stops partway through the month once the included HubSpot Credits are spent.
This article explains what Data Hub is, what each tier includes, and what changed for accounts that bought Operations Hub. It starts with the name and a comparison of two dated editions of HubSpot's catalog, then sets out the tiers and their permissions. It follows data sync and formatting fixes, programmable automation, and data studio with its credit cost on a sample account. It closes with the conditions that select Data Hub, costs and returns, a setup procedure, symptoms, common questions and the limits of the evidence.
Data sync is HubSpot's engine for connecting another app and keeping records consistent in one or both directions, available on all products and plans. A field mapping pairs a field in the other app with a HubSpot property: each app ships with defaults, and custom mappings need Data Hub. Programmable automation is HubSpot's name for the workflow actions that run code or call other systems. A dataset is a table built in data studio from HubSpot objects and external sources. HubSpot Credits are the monthly allowance that usage-based features draw on, reset each month without rollover.
HubSpot Operations Hub vs Data Hub: Name, Redirects and Catalog Changes
HubSpot's products page lists Marketing, Sales, Service, Content, Data and Revenue Hub, with Smart CRM and Agent Hub in its menu, and names no Operations Hub. HubSpot's own addresses record the rename: on 27 September 2026, hubspot.com/products/operations returned a 301 permanent redirect to the Data Hub page, hubspot.com/pricing/operations a 301 to Data Hub pricing, and the Operations Hub programmable automation page a 301 to its Data Hub counterpart. The Data Hub page's structured data gives the product the alternate name "Operations Hub", and a footnote to the Fall 2025 Spotlight post, updated 6 October 2025, states that "Data Hub replaces Operations Hub".
Whether anything beyond the name changed is answered by HubSpot's Product and Services Catalog, read against an archived copy of 31 August 2025 that still described Operations Hub. The two list identical prices: $20 a month per seat at Starter, $800 a month at Professional and $2,000 at Enterprise, with one Core Seat in each higher edition. Both place programmable automation, webhooks and data quality automation at Professional, and custom objects, the sandbox and field-level permissions at Enterprise.
The finding is in what moved, because it moved in two directions. Data quality monitoring and AI formatting recommendations, which the 2025 catalog listed only under Operations Hub Professional, now appear at Starter in every hub sold at Starter, so they no longer argue for Data Hub at all. Datasets went the other way. The 2025 tool held five datasets at Professional and, per the archived datasets article, fed custom reports. Data studio holds 200, joins external sources, and draws HubSpot Credits whenever a dataset feeds a workflow, a segment, a CRM sync or an export. Included credits rose with it, to 5,000 a month at Professional and 10,000 at Enterprise, against the 3,000 and 5,000 every hub received in 2025.
Data Hub Tiers From Free to Enterprise
The free tools include data sync, historical sync and default field mappings, and HubSpot's data sync page states that custom mappings and complex sync rules need a paid Data Hub Starter subscription. Starter adds those mappings and bulk acceptance of fixes in the data quality overview, which scans a Starter account every two weeks.
Professional is where Data Hub becomes a different product. It includes the workflows tool with three actions reserved to Data Hub, Custom code, Format data and Send a webhook, as the workflow actions list marks them. It adds webhook triggers, recurring schedules, data studio, bulk duplicate management, anomaly alerts, formatting automation rules and a real-time quality scan. Enterprise adds warehouse sources such as Amazon Redshift, larger datasets, and the governance set common to every Enterprise hub. The same step in the other hubs is set out in HubSpot Starter vs Professional.
| Capability | First Data Hub tier and documented limit | Operations Hub, August 2025 |
|---|---|---|
| CapabilityData sync with default mappings | First Data Hub tier and documented limitFree tools | Operations Hub, August 2025Unchanged |
| CapabilityCustom field mappings | First Data Hub tier and documented limitStarter | Operations Hub, August 2025Unchanged |
| CapabilityData quality overview | First Data Hub tier and documented limitStarter, scanned every two weeks; real time at Professional | Operations Hub, August 2025Professional only, as the data quality command center |
| CapabilityAI formatting recommendations | First Data Hub tier and documented limitStarter | Operations Hub, August 2025Professional only |
| CapabilityCustom code, Format data, Send a webhook | First Data Hub tier and documented limitProfessional | Operations Hub, August 2025Unchanged |
| CapabilityWebhook enrollment triggers | First Data Hub tier and documented limit10 at Professional; 100 at Enterprise | Operations Hub, August 2025Unchanged |
| CapabilityDaily, weekly and monthly scheduled triggers | First Data Hub tier and documented limitProfessional | Operations Hub, August 2025Unchanged |
| CapabilityDatasets | First Data Hub tier and documented limitProfessional, in data studio (beta); 200 per account; credits per activation | Operations Hub, August 20255 datasets of 45 columns at Professional, no credit rate |
| CapabilityCustom events | First Data Hub tier and documented limitProfessional, 10 million completions a month | Operations Hub, August 2025Enterprise only |
| CapabilityWarehouse connection | First Data Hub tier and documented limitEnterprise, as data warehouse integrations | Operations Hub, August 2025Enterprise, as Snowflake data share |
| CapabilityCustom objects | First Data Hub tier and documented limitEnterprise, 20 definitions and 1,500,000 records | Operations Hub, August 2025Unchanged |
| CapabilityIncluded HubSpot Credits | First Data Hub tier and documented limit500, 5,000 and 10,000 a month | Operations Hub, August 2025500, 3,000 and 5,000, as for every hub |
Data-Related Permissions by Tier
Data Hub's tools sit behind their own switches in the user permissions guide. Data studio carries three levels: View, which can activate a dataset for reporting; Edit; and Use, which a Super Admin must turn on before a user can spend credits by sending a dataset to a workflow, a segment, a CRM sync or an export. Data quality tool access and Data Agent access are separate switches, and building a custom code or webhook action needs Edit on workflows, with Publish to turn it on.
A custom code secret can hold a private app access token, and only a Super Admin can open private apps. Starter and Professional assign these permissions through templates, while saved permission sets, the property restrictions described in HubSpot field-level permissions and sensitive data all require Enterprise.
Data Sync, Field Mappings and Formatting Fixes
When a synced record is updated, data sync examines the entire record against its copy and updates the copy to match, and after the initial sync a change reaches the other side within ten minutes. Three settings decide which value survives: the direction, set per object as two-way or one way; the default app that overwrites the other when values disagree; and filters, which limit only which records enter the sync at first. The integrations built on this engine, including the NetSuite connector and the QuickBooks Online app, inherit all three.
The data quality overview proposes fixes for formatting issues, such as a name in capitals, which an administrator accepts one by one or, from Data Hub Starter, in bulk. Its documentation adds the rule that decides how long a fix lasts: a value later changed back by a manual update, the API, an integration, an import or a migration will "typically not be flagged again", because the tool reads the overwritten format as intended.
Read together, the two documents imply a result that neither states. Where a field arrives by a sync or a field mapping that runs one way into HubSpot, a fix accepted in HubSpot lasts only until that customer next changes in the source, when the sync restores the source value and the overview stays silent.
On the sample account priced below, 4,000 contacts arrive one way from a billing app and 1,200 carry capitalised names. If 300 of those customers are updated in the billing app during the following month, 300 fixes revert, a quarter of the work, and the overview reports none of them. HubSpot does not describe this case directly, so the setup procedure tests it on one record first.
A fix lasts when it is made where the value originates: in the source system, in a two-way mapping that carries HubSpot's corrected value back to the source, or in a Format data workflow that re-enrolls when the property changes. The field mapping rules add formatting of their own: an app with a single name field has its first word written to First Name and the remainder to Last Name.
HubSpot Data Hub Programmable Automation: Custom Code, Webhooks and Schedules
Programmable automation consists of two workflow actions and two enrollment triggers, and every one requires Data Hub Professional or Enterprise. The custom code action runs JavaScript on Node.js, or Python in beta, as a serverless function on AWS Lambda. Each execution must finish within 20 seconds and 128 MB of memory, and the secrets on one action may total 1,000 characters. When a call fails on a rate limit or a 429 or 5XX response and the code rethrows the error, HubSpot retries for up to three days, never more than eight hours apart.
The Send a webhook action sends a POST or GET request to an HTTPS address, authenticated by a request signature or an API key held as a secret, and later actions can use its response. Failed webhooks follow the same three-day schedule, except that a 4XX response other than 429 is never retried. Both actions accept a rate limit per second, minute or hour, which also delays every later action in the workflow.
The two triggers bring enrollment from outside the record. A webhook trigger enrolls records when an external system sends a webhook, and the catalog allows 10 at Professional and 100 at Enterprise. The Based on a schedule trigger runs once or annually in any hub that includes workflows, but recurs daily, weekly or monthly only with Data Hub Professional or Enterprise. Data Hub Professional includes the workflows tool, so no other Professional hub is needed for any of this.
One limit concerns maintenance: the workflow editor can undo changes from the previous 30 days except those to custom code actions, so each action's code needs a copy in version control.
Data Studio Datasets and Their Credit Cost on a Sample Account
Data studio, in beta with Data Hub Professional or Enterprise, builds a dataset from up to five HubSpot or external sources, such as Google Sheets, files and data sync apps. The catalog allows 30 formula fields per dataset at Professional and 50 at Enterprise, and 10 million stored external rows against 30 million. A report can use a dataset at no credit cost. A workflow, a segment, a CRM sync or an export costs 25 credits per use when the primary source holds fewer than 500,000 rows, 75 up to five million and 200 above that.
The sample account is sample data and describes no client. It runs Sales Hub Professional for twelve users and adds Data Hub Professional at the catalog's $800 a month, $9,600 a year, with no required onboarding listed; the pricing page showed $720 a month for one billing option. Included credits are not additive across products, so the account moves from 3,000 to 5,000 a month, not to 8,000. It joins an 18,000-row product usage sheet to its companies and syncs the result to a company property. The catalog does not say whether each scheduled run counts as one use, so the table takes each run as one use and the setup procedure checks that reading.
| Sync frequency | Runs and credits a month | Beyond 5,000 included credits |
|---|---|---|
| Sync frequencyDaily | Runs and credits a month30 runs, 750 credits | Beyond 5,000 included creditsNone; 4,250 credits remain |
| Sync frequencyHourly | Runs and credits a month720 runs, 18,000 credits | Beyond 5,000 included credits13,000 credits, $130 a month |
| Sync frequencyEvery 15 minutes | Runs and credits a month2,880 runs, 72,000 credits | Beyond 5,000 included credits67,000 credits, $670 a month |
At $0.010 a credit, the hourly setting takes the year from $9,600 to $11,160, and the fifteen-minute setting takes it to $17,640, 84% above the licence. Two defaults in the credits documentation decide when the sync stops rather than bills. Without purchased credits, usage-based features pause once the included credits are spent, which at 25 a run comes after 200 runs: during the ninth day at hourly and the third at fifteen minutes. Buying the capacity pack that overages require sets a spend limit 50% above the new total, 9,000 credits against 6,000, which halts an hourly sync after 360 runs, at the end of the fifteenth day.
Data Hub Selection Against Other Hubs and Alternatives
According to the catalog, Marketing, Sales and Service Hub Professional already include the workflows tool, 40 calculated properties, association labels, duplicate management and a daily data quality scan; the limits of the duplicate tools are set out in HubSpot duplicate management. All of it reads and writes one database, HubSpot Smart CRM, whichever hub holds the seat.
Five needs select Data Hub: a field mapping beyond an app's defaults, at Starter; code or an HTTP call inside a workflow; enrollment from an external event or on a recurring schedule; external data joined and written back to records; and, at Enterprise, a warehouse source. A requirement list with none of these does not need Data Hub, because the quality overview at Starter and duplicate management at Professional already handle record-by-record repair.
Two alternatives win on specific axes. An external function called through a private app wins on runtime and change control: it is not bound by the 20-second and 128 MB limits, and it can live in version control with tests. A native connector or an existing middleware platform wins on cost where the need is one or two flows, since $9,600 a year buys the whole programmable surface whether one action uses it or forty. Data Hub wins where the logic belongs to a record's workflow: enrollment history, retries and action logs stay in HubSpot, and the data passes through no third system. The catalog also lists six-hour HubSpot consulting services, one to launch a single coded-action process and one to enable data sync for one system.
Costs and Returns of Data Hub Professional
The firm implements HubSpot as a Solutions Partner, and a Data Hub Professional account carries more configuration work for a partner than one that stays on default mappings; the conditions below should be read with that interest in view.
What Data Hub Professional buys is the ability to change data inside the workflow that already governs a record: normalising a value, computing from an external lookup, notifying a system with no native integration, and enrolling records from outside events. What it costs is $9,600 a year at list, the credits that dataset activations draw, and a maintenance duty, since every custom code action is software someone must own and test.
The case is strongest where several systems write to the same records and rules must apply at the point of write, where dozens of flows would otherwise sit in middleware, and where external data must drive segments and workflows. It is weakest for an account whose native connectors map what it needs, whose data problem is a one-off cleanup, or whose only requirement is one outbound notification that an existing middleware flow already carries.
HubSpot Operations Hub Setup Procedure
- Write each data requirement as a sync, a format, a computation, a call or a dataset, and mark its first tier from the table above. Any requirement marked Professional sets the edition.
- In Settings, open Users & Teams, give builders View, Edit and Publish on workflows and View and Edit in data studio, and grant data studio's Use permission only to users who may spend credits.
- Configure each sync from Settings > Integrations > Connected Apps: the direction for each object, the default app for conflicts, any custom mappings, and filters, which govern only the initial sync.
- Test one record before any bulk fix. Accept one fix under Data Management > Data Quality > Formatting issues, change a different field on that customer in the source app, wait ten minutes, and read the property's history in HubSpot.
- Build each custom code action from the Data ops section of the workflow editor, add its secret and outputs, set a rate limit, and click Test action, whose log shows runtime and memory.
- Build each dataset in Data Management > Data Studio, choose Use in > Sync to CRM, and set a match key, the lowest frequency the process tolerates, and a credit limit for data studio.
- Verify after 24 hours. Under Account & Billing > HubSpot Credits, compare data studio's consumption with runs multiplied by the per-use rate, then open the dataset's View sync health and one enrolled record's custom code log.
Data Hub Symptoms, Causes and Fixes
Names fixed in the data quality overview are capitalised again, and nothing is flagged. The one-way sync restored the source value at the customer's next update, and the overview does not re-flag values an integration writes, so the fix belongs in the source or in a two-way mapping.
A data studio CRM sync stops partway through the month with no error in the dataset. The included credits ran out, or purchased credits reached the spend limit, and usage-based features pause until the reset date.
The custom mapping panel offers only Try out custom mappings, because the account holds no paid Data Hub subscription.
A webhook action stops after one failed attempt. The endpoint returned a 4XX status other than 429, which workflows never retry, so the request itself needs correcting.
A custom code action fails on large enrollments. It exceeded 20 seconds or 128 MB, or the code caught a 429 from HubSpot's API without rethrowing it, which prevents the retry.
Frequently Asked Questions
Was HubSpot Operations Hub Renamed Data Hub?
Yes, on the evidence of HubSpot's own site. The Operations Hub product, pricing and programmable automation addresses redirect permanently to their Data Hub counterparts, the Data Hub page lists Operations Hub as its alternate name, and a footnote to the Fall 2025 Spotlight post reads "Data Hub replaces Operations Hub".
Is HubSpot Operations Hub Pricing Different Under the Data Hub Name?
Not at list price, which is $20 a seat, $800 and $2,000 a month in both catalogs. Data Hub Professional now includes 5,000 credits a month, and data studio activations draw on them.
Does HubSpot Operations Hub Setup Require a Developer?
Only for custom code and for the systems a webhook calls. Data sync, custom mappings, Format data and data studio are configured in the interface.
Is HubSpot Data Hub Programmable Automation Available on Starter?
No. Custom code, Send a webhook, webhook triggers and Format data all require Data Hub Professional or Enterprise.
Which Differences Matter in HubSpot Operations Hub vs Native CRM Integrations?
A native integration built on data sync is free with its default mappings. Data Hub adds custom mappings at Starter, then code, webhooks and datasets at Professional, once a field or a system falls outside what the connector maps.
Boundaries of the Evidence, September 2026
Every tier gate, limit and price here is as documented in September 2026, read on 27 September 2026 from the catalog, the knowledge base, the developer documentation and the pricing page. The comparison with Operations Hub rests on Wayback Machine captures of the catalog and the datasets article, because HubSpot publishes no version history. The pricing page and the catalog disagree on the Professional scan frequency, two weeks against real time, and this article follows the catalog. It does not cover the data agent, Breeze or Smart CRM's editions.
The evidence has three further limits. The overwrite of a fixed value by a one-way sync is an inference from two documents, and the procedure's fourth step is its test. The credit arithmetic assumes each scheduled run is one use and that no other feature draws credits. Data studio and the quality overview are in beta, and the credits documentation warns that beta rates may change; nothing here was retested in a live account.
In Summary
Data Hub is Operations Hub renamed, as HubSpot's permanent redirects show, and the catalog lists it at the same three prices. Data sync with default mappings is free, custom mappings start at Starter, code, formatting, webhooks and recurring schedules start at Professional, and warehouse sources start at Enterprise.
Two changes matter to an Operations Hub account. Data quality monitoring now comes with any Starter hub, which removes it from the case for Data Hub, and datasets became data studio, where every activation outside a report draws credits, so a sync's frequency is a pricing decision.
The consequence is an order of work: decide which system owns each field before fixing a value in HubSpot, since a fix downstream of a one-way sync lasts until the source next changes, and read the credits tab the day after a dataset sync starts, not at the first invoice.