CRM
HubSpot Sales Hub
Rest of world Report an error0–5 in half steps. 5 means the rubric's top anchor is met on the evidence.
by HubSpot, Inc. · www.hubspot.com
Compare with Pipedrive → Compare with Zoho CRM → Compare with Microsoft Dynamics 365 Sales → Report an error on this page Is this your product? →
Read this page as one judge. Each weighs the same scores by what they care about.
The panel's verdict
HubSpot Sales Hub is judged strongest on pipeline & forecasting: pipelines scale 2/15/100 across Starter/Professional/Enterprise, forecasting is default-plus-custom with team hierarchies at Enterprise, and dashboards reach 100 at 50 reports each — though judges dock it for unevidenced stage gates, required fields and quota tracking. Integrations sits at 4-5 on a 2,000+ app marketplace with no evidenced API documentation, webhooks or two-way email/calendar sync. Everything else is thin: automation 2-3, contact & data model 2-4, sovereignty 0-1, import, export & exit 0-1. The clearest split is contact & data model, where the data protection officer read tiered pipelines and smart-properties runs as a real relational model (4) while the others scored silence on dedupe and merge as absence (2). No flagged split passed the threshold. No currency price for any tier and no credit price for the AI agents appears in the evidence, so no invoice is computable; the vendor is HubSpot, Inc. of Cambridge MA (NYSE: HUBS), with residency, ownership, jurisdiction and subprocessor exposure all marked unknown.
Speaks for it
- Deal pipelines scale 2/15/100 across Starter/Professional/Enterprise, and pipeline & forecasting is the product's highest-scoring criterion.
- Default and custom forecasting at Professional, extending across team hierarchies at Enterprise.
- Publishes granular per-tier limits — calling minutes 500–12,000, dashboards 10–100, transcription hours — credited as honest limit disclosure.
- A 2,000+ app marketplace is credited as a genuine directory.
- View-only seats are available at no additional cost under the per-seat model.
Held against it
- No currency figure for any tier and no credit price for the AI agents appear in the evidence, so no invoice can be computed.
- The only evidenced automation is credit-metered AI agents — 10 credits per smart-properties run, 50 per resolved conversation, 100 per recommended outreach — with no workflow builder, sequences or routing.
- No captured page evidences dedupe, merge, custom fields or field types (contact & data model 2-4).
- Legal jurisdiction, ownership, data residency and subprocessor exposure are each marked unknown in the evidence; the vendor is HubSpot, Inc., Cambridge MA, NYSE: HUBS.
- Import, export and migration are unevidenced in both directions; deletion on request is the only lifecycle control (import, export & exit 0-1).
Best for
- You run many parallel sales processes and need tiered pipelines (up to 100 at Enterprise) plus forecasting across team hierarchies.
- You want a broad native app directory (2,000+ apps) and don't depend on documented APIs or webhooks.
- You have many read-only stakeholders, since view-only seats are available at no additional cost.
- You need multi-language support — English, French, German, Japanese and Portuguese are named.
Avoid if
- You must know your invoice before committing — ask the vendor: the public pages we read do not show it
- Your compliance review requires a DPA or subprocessor list — neither appears in the evidence, and each sovereignty attribute is marked unknown.
- Your team relies on workflow automation — ask the vendor: the public pages we read do not show it
- You need a documented exit path — ask the vendor: the public pages we read do not show it
The scores
Contact & data model
Show reasoningHide reasoning
How this is scored
Depth and hygiene of the core records — contacts, companies, deals, custom fields, deduplication.
0 — A flat address book; no relations between people, companies and deals.
3 — The standard three objects exist but customization is cosmetic: few field types, no required fields, duplicates accumulate unchecked.
5 — Custom fields across objects, basic dedupe on create, activity history on the record; complex models still need workarounds.
8 — Rich field types with validation, merge and dedupe tooling, relationships beyond the standard model, permissioned field access.
10 — A genuinely modelable system: custom objects or equivalent, bulk hygiene tooling, field-level history — the data model bends to the business, not the reverse.
The Solo Seller
The pricing page proves deals with tiered pipeline counts and property-based lead scoring, but nothing here evidences company objects, custom field types, required fields, or any dedupe/merge tooling — the core record story is a blank, and blank means my contact hygiene is unverified. 2
The Data Protection Officer
Deals-as-pipelines up to 100 and 'smart properties' runs tell me a real relational model with custom properties exists, but the evidence is silent on deduplication, merge, field validation and permissioned field access — for a database of other people's personal data, unevidenced hygiene is unevidenced hygiene. Slightly above the cosmetic-customization floor only because the pipeline model is clearly substantial. 2
The Sales Ops Lead
The only record-structure facts are deal-pipeline counts and lead-score limits; nothing evidences custom fields, required fields, dedupe or merge tooling. For a criterion decided by hygiene, silence is the answer — I won't put a 40-seat org on an unevidenced data model. 2
The RevOps Integrator
Deals and pipelines are clearly real (up to 100 pipelines) and lead scoring implies property infrastructure, but the evidence is silent on custom fields, dedupe, merge, and anything beyond the standard objects — I won't wire a revenue stack to a data model nobody evidenced. 2
The Migrator
The evidence never shows me a contact, a company object, a custom field, or a dedupe button — the closest it comes is deal pipelines and lead scoring, which proves deals and scoring properties exist and nothing about hygiene. No evidenced dedupe or merge means duplicates accumulate unchecked as far as this record goes, and I score silence as absence. 2
The Skeptic
The captures prove deals, pipelines and lead scores exist, but not one word on field types, required fields, dedupe, merge or custom objects — the hygiene layer that makes a CRM trustworthy is simply absent from every page captured. Silence on data hygiene is not neutrality, it's usually where the mess lives. 2
Pipeline & forecasting
Show reasoningHide reasoning
How this is scored
Whether the pipeline enforces a process and produces reporting a finance department accepts.
0 — A kanban of deals with amounts; nothing is enforced, nothing rolls up.
3 — Multiple pipelines and stage probabilities exist, but no required fields per stage and only canned reports.
5 — Stage requirements, win/loss capture and standard forecast views; custom reporting is limited or an add-on.
8 — Enforceable process (required fields, stage gates, routing), custom report builder, forecast by team and period with quota tracking.
10 — Forecasting a CFO signs: weighted and commit-based views, historical snapshots, attainment tracking, and reporting composable enough to answer questions the vendor never anticipated.
The Solo Seller
Multiple pipelines (2 on Starter, 15 on Professional) and custom forecasting with up to 75 dashboards at Professional are real, but forecasting starts two tiers above Starter and nothing evidences required fields, stage gates or quota tracking — at my one seat I'd get a kanban with counts, not an enforced process. 2
The Data Protection Officer
Default and custom forecasting with team-hierarchy rollups and up to 100 dashboards at 50 reports each sits above canned reporting, but nothing in the evidence evidences required fields per stage, stage gates, quota attainment or forecast snapshots. The enforcement layer a finance department would sign off on is invisible, so I stop just past the mid anchor. 2
The Sales Ops Lead
Pipelines scale by tier (2/15/100) and forecasting is real — default plus custom, across team hierarchies at Enterprise. But there is zero evidence of required fields per stage, stage gates, quota attainment or forecast snapshots, and dashboards are hard-capped per tier (10/30/75/100), so this stops well short of CFO-grade. 2
The RevOps Integrator
Pipelines scale to 100 and forecasting goes default-plus-custom with team hierarchies at Enterprise, plus 75–100 dashboards — but nothing evidences required fields per stage, stage gates, quota tracking or historical snapshots, so the reporting exists while the enforcement doesn't show up. 2
The Migrator
Up to 100 pipelines and custom forecasting across team hierarchies at Enterprise are real scaffolding, but the evidence says nothing about required fields per stage, stage gates, win/loss capture, or quota tracking. That lands between 'pipelines exist, nothing enforced' and 'stage requirements' — closer to the former, since no enforcement is evidenced at all. 2
The Skeptic
Default and custom forecasting is real and pipeline counts scale by tier (2/15/100), but nothing evidences stage gates, required fields per stage, win/loss capture or quota tracking — the enforcement layer a CFO asks about is unevidenced. And the pipeline count itself is an upsell lever: your second pipeline already costs you. 2
Automation & workflows
Show reasoningHide reasoning
How this is scored
Triggers, sequences and routing that remove manual work without a consultant.
0 — No automation beyond email templates.
3 — Simple if-this-then-that on records; no multi-step sequences, no branching.
5 — Multi-step workflows with delays and branching, email sequences, basic lead routing — capped by tier or count.
8 — Full workflow builder across objects with webhooks as steps, round-robin routing, sequence analytics.
10 — Automation as infrastructure: versioned workflows, error handling, per-step logs, and limits generous enough that the feature is not its own upsell.
The Solo Seller
Snippets and meeting links are templates, and the only automation evidenced is credit-metered AI agents (100 credits per recommended outreach, 50 per resolved conversation) with no workflow builder, triggers, sequences or routing anywhere in the evidence — automation here is an upsell with an unreadable meter, not infrastructure. 2
The Data Protection Officer
The only automation evidenced is credit-metered AI agents — 50 credits per resolved conversation, 100 per recommended outreach, 10 per smart-properties run — plus tracked-term conversation intelligence, with no workflow builder, branching, sequences or routing anywhere in the evidence. Automation sold by the drop, with the meter itself unpriced, is precisely the feature-as-upsell pattern the top anchor warns against, and the prospecting agent smells of automated outreach to scraped leads. 2
The Sales Ops Lead
Snippets, lead scores and credit-metered AI agents exist, which is trigger-adjacent at best. No workflow builder, sequences, branching or routing in evidence — and agent-driven actions bill in credits whose cost is never stated, which smells like automation as its own upsell. 2
The RevOps Integrator
The only automation evidenced is credit-metered AI agents (10–100 credits per run) plus snippets and meeting links; no workflow builder, sequences, branching or webhooks appear anywhere. Automation here is sold as an upsell unit, not infrastructure. 2
The Migrator
The only automated work evidenced is the credit-metered AI agents — 50 credits per resolved conversation, 100 per prospecting outreach — plus snippets and templates. No workflow builder, multi-step sequences, branching, or routing appear anywhere, and the per-run metering is precisely automation sold as its own upsell. 2
The Skeptic
No workflow builder, sequences or routing appear anywhere in the evidence; the only automation evidenced is AI agents metered in credits — 10 per smart-properties run, 50 per resolved conversation, 100 per recommended outreach — with no credit price in evidence. Automation here is not a feature, it's a meter running. 2
Integrations & API
Show reasoningHide reasoning
How this is scored
Two-way email/calendar sync, native connections, and an API that treats external systems as first-class.
0 — No API, no sync; data enters by CSV and leaves never.
3 — One-way email logging (BCC), a handful of Zapier triggers, a read-mostly API.
5 — Two-way email and calendar sync for the major providers, a documented REST API with sane limits, a modest native directory.
8 — Deep native integrations for the revenue stack, webhooks, OAuth apps, API parity with the UI for core objects.
10 — An ecosystem: public API covering everything the UI does, event streams, a marketplace, and sync so reliable the CRM can be the system of record without being the only system.
The Solo Seller
A 2,000+ app marketplace is a genuinely serious directory, but the evidence never shows two-way email/calendar sync or a documented API, so I can't confirm the CRM even reads my inbox — the integration I'd need on day one is unevidenced. 2
The Data Protection Officer
A marketplace of 2,000+ apps and web services is a genuine ecosystem signal, but the evidence evidences no two-way email or calendar sync, no API documentation, no webhooks and no sync guarantees. I can credit a large directory; I cannot credit depth or API first-classness the evidence does not show. 2
The Sales Ops Lead
A 2,000+ app marketplace is a serious native directory. But the evidence says nothing about email/calendar sync, a REST API, webhooks or OAuth — the 'external systems as first-class' half of this criterion has no evidence at all. 2
The RevOps Integrator
A 2,000+ app marketplace plus native calling and meeting links is a genuine directory, but the evidence says nothing about two-way email/calendar sync, REST API docs, or webhooks — and without evidenced webhooks I cannot treat external systems as first-class. 2
The Migrator
A 2,000+ app marketplace is a big native directory, but the evidence is silent on the API, two-way email/calendar sync, webhooks, and limits — and for me an unevidenced API is an unevidenced exit route. Directory without documented API access stays below rubric level 5. 2
The Skeptic
'Over 2,000 leading apps and web services' is a count, not a capability — no evidence of two-way email/calendar sync, no API documentation, no webhooks, nothing on OAuth or parity. A marketplace number is the easiest thing on a pricing page to say and the hardest to hold to. 2
European sovereignty
panel opinion
Show reasoningHide reasoning
How this is scored
Where other people's personal data actually lives and under whose law — entity, hosting, subprocessors, DPA.
0 — US entity, US-default hosting, no public DPA or subprocessor list.
3 — A DPA exists and an EU region is available on request or on top tiers; subprocessor exposure to US CLOUD Act reach is broad and undocumented.
5 — EU hosting selectable at signup, published DPA and subprocessor list; core processing still touches US entities.
8 — EU entity or EU-default hosting with published subprocessors, consent-friendly fields, retention and deletion controls; residual US exposure named and narrow.
10 — European entity, EU-only processing including subprocessors, deletion that provably deletes — sovereignty a DPO can sign off without a caveat file.
The Solo Seller
A US-listed Cambridge MA vendor with every sovereignty attribute unconfirmed, no DPA or subprocessor list on any captured page, and only a privacy policy promising deletion-on-request and retention handling — my contacts' personal data sits under undocumented third-party reach. 1 3
The Data Protection Officer
A US entity (Cambridge MA, NYSE-listed) with data residency, ownership and subprocessor exposure all marked unknown, no DPA and no subprocessor list in evidence — the deletion-on-request clause and the hedge-and-isolate retention wording in the privacy policy are the entirety of the GDPR posture, and a Berlin sales phone line is not an EU data center. Add 750–1,500 hours of call transcription per month processing third parties' personal data with zero evidenced European controls: nothing here a DPO could sign. 3 2
The Sales Ops Lead
US-incorporated, publicly listed vendor, every sovereignty attribute unconfirmed, and no DPA or subprocessor list in evidence. The privacy policy's delete-on-request and retention-until-deletion language is the only positive, and it says nothing about where my customers' personal data actually lives. 3
The RevOps Integrator
US entity (Cambridge MA) with residency, ownership and subprocessors all unknown — the evidence's own computed sovereignty score is 0/10, and a privacy-policy deletion clause doesn't change whose law the data sits under or who else can reach it. 3
The Migrator
The vendor is HubSpot, Inc. of Cambridge MA — a US entity — and the captured privacy policy offers deletion and retention language but no DPA, no subprocessor list, and no EU hosting; the German sales phone line is a sales office, not data residency. That is rubric level 0 exactly. 3 2
The Skeptic
US entity (Cambridge MA, NYSE: HUBS), data residency unknown, no DPA and no subprocessor list anywhere in evidence — the exact 0-anchor profile, and the computed sovereignty score agrees. A privacy policy promising deletion tells a DPO nothing about whose law the data sits under while it lives. 3
Import, export & exit
Show reasoningHide reasoning
How this is scored
Getting in with history intact, and out with everything — the anti-lock-in criterion.
0 — Import is a support ticket; export is contacts-only CSV missing notes, activities and files.
3 — Self-serve CSV import with mapping; export covers main objects but drops activity history, attachments or relations.
5 — Guided migration from major competitors, full-object CSV export, API access sufficient to rebuild the database elsewhere with effort.
8 — Complete export (all objects, history, files) self-serve in open formats, import with dedupe and dry-run, no API metering that prices exit out of reach.
10 — Exit as a feature: documented full-fidelity export, migration tooling both directions, and contractual data-return terms — leaving is a weekend, which is exactly why you can stay.
The Solo Seller
Nothing in the product or pricing pages evidences import or export of any kind, not even a CSV mention; the only exit-adjacent fact is the privacy policy's deletion promise, which is a lawyer's exit, not a data exit. 1 3 2
The Data Protection Officer
The evidence is entirely silent on import, export, migration and API data-return — no CSV, no full-fidelity export, nothing an exit plan could rest on; the only lifecycle control evidenced anywhere is deletion on request. By the anchors, silence on exit scores as lock-in, and lock-in on a database of other people's data is a compliance problem, not just a commercial one. 3
The Sales Ops Lead
Nothing evidences import, export or migration in either direction — no CSV tooling, no migration path, no API mention anywhere in the evidence. The only data-lifecycle facts are deletion on request and retention isolation, which is about removing data, not returning it. 3
The RevOps Integrator
Total silence: no import, export, or migration capability appears anywhere in the registry — not even the privacy policy, which covers deletion but not portability. Unproven exit paths score at the bottom of the anchors, and I assume lock-in until a vendor shows me the door. 3 2
The Migrator
Nothing in this sheet evidences import, export, or migration tooling — no self-serve CSV, no API route out, no data-return terms. The only exit-relevant text is a promise to delete or anonymize on request, which destroys my database rather than returning it, and I score that absence as the bottom of the scale. 3
Pricing transparency
Show reasoningHide reasoning
How this is scored
Whether a buyer can compute the real invoice — per seat, per month, at renewal — from public pages alone.
0 — No public prices at all; every tier is a sales conversation.
3 — An entry price exists, but the tiers most buyers need are unpriced, or per-seat maths is obscured by bundles, minimums or mandatory onboarding fees.
5 — Most tiers carry real numbers with billing period and VAT treatment stated, but at least one commonly needed capability hides in an unpriced tier or usage-metered add-on.
8 — Every tier priced publicly including annual/monthly split; limits, seat minimums and overage rules stated; only genuine enterprise contracts are custom.
10 — Complete price computability: every tier, add-on, usage meter and renewal rule public, with a calculator or table that makes the invoice at 3 and 30 seats a two-minute exercise.
The Solo Seller
The per-seat model, free view-only seats and exhaustively public limit tables (calling minutes, pipelines, dashboards) are commendable, but no actual price figures appear in evidence and the AI features bill in credits whose price never shows up — I can read the menu but cannot compute the invoice on my own card. 2 2
The Data Protection Officer
The pricing page does publish hard per-tier limits — 500 to 12,000 calling minutes, 1 to 100 pipelines, transcription hours — plus the per-seat model and free view-only seats, which is honest limit disclosure. But no currency figure for any tier appears in the evidence, and AI capabilities are metered in credits whose price is never stated, so the real invoice cannot be computed from this evidence. 2
The Sales Ops Lead
The per-seat model, free view-only seats and every tier's limits are published, which is good scaffolding — but not one currency figure, billing-period split or VAT treatment appears, and AI agents are metered in credits with no stated credit price. I cannot compute a 40-seat invoice from this evidence, so real price transparency is unproven. 2
The RevOps Integrator
Limits are granular (calling minutes, pipelines, dashboards, seats, view-only seats free) but not one currency figure is confirmed, and the credit metering behind the AI agents has no public per-credit price — no buyer can compute a real invoice from this sheet. 2 2
The Migrator
The pricing page enumerates editions, a per-seat model, tier limits, and even credit costs per agent run, but not one currency figure, billing period, or VAT treatment made it into the record — so no invoice is computable from this evidence. The AI credit meter is priced per run without a credit price, which is the classic un-computable line item. 2
The Skeptic
The pricing pages captured publish generous limits (calling minutes, transcription hours, dashboards) yet contain no actual per-seat price for any tier, so the invoice cannot be computed from the evidence at all. Worse, the AI features run on a credit meter with per-action costs stated but no credit price, bundle size or overage rule — the textbook unpriced usage add-on waiting to detonate at renewal. 2 2
European sovereignty — proven facts
0 of 4 dimensions provenBuilt only from facts shown on the vendor's own pages. A dimension we could not prove is left open, not scored as zero.
| Legal entity | Not determined | — | uncited Report an error |
|---|---|---|---|
| Ownership | Not determined ⚠ unverified | — | uncited Report an error |
| Data residency | Not determined | — | uncited Report an error |
| Subprocessors | Not determined ⚠ unverified | — | uncited Report an error |
Where this could be wrong
- Evidence ages. The oldest capture behind this page is from 11 Sep 2026. Vendors change pricing and policies without notice; every fact reflects its source as of the capture date shown in the registry.
- Weak sourcing — Ownership, Subprocessors. Not confirmed on the vendor’s own pages as captured.
- AI can misread a source. Extraction and judgement are automated; a citation guarantees traceability, not infallibility. If something here is wrong, say so — no account needed, every report is decided within 5 business days, and accepted corrections are published.
What we left out
A claim that does not survive our checks costs us the claim, not the page. This is what was taken off this one.
- 25 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material. Know more? Tell us
- 16 pricing facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material. Know more? Tell us
- 1 data fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material. Know more? Tell us
- 1 integrations fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material. Know more? Tell us
- 1 support fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material. Know more? Tell us
- 2 sovereignty dimensions could not be confirmed on the vendor’s own pages and are shown as unknown. Know more? Tell us
- 6 of the readings below were written against an earlier fact sheet — a fact has been corrected, added or pulled since. Until the panel next runs on this product you are reading the older judgement. Know more? Tell us
Sources (13)
The pages every claim on this page was read from — each one checked, dated, and kept verifiable.
- 1 Vendor product page www.hubspot.com Checked 15 Sep 2026 Details →
- 2 Vendor pricing page www.hubspot.com Checked 15 Sep 2026 +1 earlier capture: 11 Sep 2026 Details →
- 3 Privacy policy legal.hubspot.com Checked 15 Sep 2026 Details →
- 4 Security / trust page legal.hubspot.com Checked 30 Sep 2026 Details →
- 5 Contact & data model — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 6 Contact & data model — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 7 Pipeline & forecasting — found from sitemap www.hubspot.com Checked 1 Oct 2026 Details →
- 8 Pipeline & forecasting — found from sitemap www.hubspot.com Checked 1 Oct 2026 Details →
- 9 Automation & workflows — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 10 Automation & workflows — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 11 Integrations & API — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 12 Import, export & exit — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →
- 13 Import, export & exit — found from sitemap knowledge.hubspot.com Checked 1 Oct 2026 Details →