top of page

Claude Team vs Claude Enterprise: Complete Comparison and Report on Seat Prices, Usage-Based Billing, Admin Controls, and Compliance

13 minutes ago
14 min read
Claude logos with the title Claude Team vs Claude Enterprise comparison of seat prices, usage-based billing, admin controls and compliance, Data Studios original cover

Claude Team and Claude Enterprise are Anthropic's two organization plans for Claude, and in November 2026 they run on two different economic models. Team is a seat plan for 2 to 150 people with usage limits built into each seat: a Standard seat costs $20 per month billed annually ($25 monthly) and a Premium seat $100 per month billed annually ($125 monthly), with every seat including Claude Code and Cowork. Enterprise, sold self-serve since February 12, 2026 and also through Anthropic's sales team, charges $20 per seat per month on an annual contract with a 20-seat minimum, but the seat fee covers access only: every prompt, Claude Code session and Cowork task is billed on top at standard Claude API rates, with no per-seat usage limits and no included token allowance.


The verdict follows from that split. Claude Team is the cheaper and simpler plan for organizations under 20 people and for any group whose heavy users fit inside Premium-seat limits, because the price is fixed and known in advance. Claude Enterprise is the plan for organizations that need SCIM provisioning, audit logs, the Compliance API, custom data retention, role-based access, IP allowlisting or a HIPAA-ready configuration, since Anthropic's pricing table marks every one of those as unavailable on Team. On cost, the dividing line is about $80 of API usage per person per month: below that amount an Enterprise seat plus usage costs less than a Team Premium seat, above it Enterprise costs more and has no usage ceiling unless administrators set spend limits.


Both plans share the same core product: the Claude apps, Claude Code, connectors, enterprise search, SSO, usage analytics and no model training on customer content by default. The differences sit in billing, identity management, logging and regulated-data controls. All prices and plan terms below were verified on Anthropic's official pricing page, Claude Help Center articles and Claude Platform documentation on November 21, 2026.


··········


CLAUDE TEAM AND CLAUDE ENTERPRISE: HOW THE TWO BUSINESS PLANS ARE BUILT.


Claude Team is a self-serve plan bought at claude.ai/upgrade. Anthropic's Help Center describes it as a plan for a minimum of two members that supports up to 150 seats, and the pricing page presents it as "For teams of 2 to 150". Each seat is assigned to one named person, cannot be shared, and is one of two types, Standard or Premium, which can be mixed in the same organization. Anthropic's January 28, 2026 post on Team updates states that Claude Code is included with every seat, and the current Team card on the pricing page lists Claude Code and Claude Cowork side by side.


Claude Enterprise was reorganized during 2026 around a single "Enterprise seat". According to the Help Center article on Enterprise billing, dated October 13, 2026, the seat fee "covers platform access but no usage", and all usage across Claude, Claude Code and Cowork is billed separately at standard API rates. Older Enterprise contracts used other seat structures ("Chat" and "Chat + Claude Code" seats, or "Standard" and "Premium" seats with a flat fee that included a usage allowance); Anthropic says these legacy seats are not available for new contracts and move to the usage-based model at renewal. Organizations that enable the HIPAA-ready configuration are the exception and keep separate Chat and Chat + Claude Code seats.


Anthropic's Help Center summarizes the relationship in one line: Enterprise "includes everything in the Team plan, plus additional security and compliance features." In practice the plans diverge on three axes: how usage is paid for, how users and data are governed, and how the plan is purchased and invoiced. A broader overview of Anthropic's consumer and team tiers is in our guide to Claude pricing and plan limits.


··········


CLAUDE TEAM VS CLAUDE ENTERPRISE SEAT PRICES AND BILLING TERMS.


The headline seat prices look identical at $20, which hides the most important difference. A Team Standard seat at $20 includes a usage allowance; an Enterprise seat at $20 includes none, so its real monthly cost is the seat fee plus whatever the user consumes at API rates. Team can be paid monthly or annually by credit card only. Self-serve Enterprise is annual only and accepts credit card or ACH bank transfer, while sales-assisted Enterprise adds invoicing with net terms.


........


Item

Claude Team

Claude Enterprise

Seat price

Standard $20/mo annual, $25 monthly; Premium $100/mo annual, $125 monthly

$20 per seat per month, billed annually

What the seat includes

Usage allowance per seat (5-hour and weekly limits)

Access only; all usage billed at API rates

Seat range

2 to 150 seats

Minimum 20 seats (Help Center: 50 for sales-assisted)

Billing cycle

Monthly or annual

Annual only

Payment methods

Credit card only

Self-serve: card or ACH; sales-assisted: card, ACH, invoice

Pooled API credits

Up to $500 per month

Not included

Extra usage

Usage credits at standard API rates

Usage is the bill; prepaid credits (self-serve) or monthly in arrears (sales)

How to buy

Self-serve at claude.ai/upgrade

Self-serve at claude.ai/create/enterprise or through sales


........


Seat changes follow similar rules on both plans. New seats are prorated for the rest of the billing cycle and charged immediately, and removing a member frees the seat without refunding it. On Team, a reduction in the seat total takes effect at renewal with no prorated refund, and only one scheduled seat change can be pending at a time. On self-serve Enterprise, seats cannot be removed during the annual term at all; sales-assisted customers change seats through their account manager. For sales-assisted Enterprise invoices of $50,000 or more, Anthropic accepts bank transfer (ACH or wire) only.


The seat-minimum figure is the first inconsistency in Anthropic's own pages. The pricing page, the Enterprise solutions page and the Enterprise seats article all state a 20-seat minimum, while the "What is the Enterprise plan?" article, dated September 1, 2026, lists 20 seats for self-serve and 50 seats for sales-assisted contracts. Organizations planning a sales-assisted purchase should confirm the minimum with Anthropic before budgeting.


··········


HOW CLAUDE ENTERPRISE USAGE-BASED BILLING WORKS AT API RATES.


On a usage-based Enterprise plan, every message is priced like an API call to the model that answered it. The Help Center states that there are "no plan or seat-level usage limits" and "no included token allowance", and that administrators control cost with spend limits at the organization and individual user levels. Owners can set those limits to "unlimited", in which case usage is still billed. On self-serve plans, usage is prepaid in USD credits shared across the organization and stops when the credits run out; on sales-assisted plans it is billed monthly in arrears.


The rates are the public Claude API rates. The table lists the four models most relevant to business use, as published in the Claude Platform documentation on November 21, 2026.


........


Model (API)

Input per 1M tokens

Output per 1M tokens

Cache read per 1M

5-minute cache write per 1M

Context / max output

Claude Fable 5.2

$8.00

$40.00

2.5% of input (no dollar figure on overview)

Not listed on overview

1M / 128K

Claude Opus 5.5

$4.00

$20.00

$0.20

$5.00

1M / 128K

Claude Sonnet 5.5

$2.00

$10.00

$0.10

$2.50

1M / 128K

Claude Haiku 5.5 (prompts up to 100K)

$0.10

$0.50

$0.01

$0.125

1M / 128K

Claude Haiku 5.5 (prompts over 100K)

$0.50

$2.50

$0.05

$0.625

1M / 128K


........


Two billing details change the arithmetic. If an Enterprise organization turns on US-only inference, Anthropic bills usage at 1.1 times standard API rates for Claude Opus 4.6, Claude Sonnet 4.6 and later models. And because model choice drives cost directly, Enterprise administrators get controls that Team does not have for this purpose: Anthropic's Claude Code for Enterprise page says Enterprise admins can set spend limits for the organization, for groups or for individuals, choose which models people can use, and cap effort levels by role.


··········


CLAUDE TEAM USAGE LIMITS FOR STANDARD AND PREMIUM SEATS.


Team seats are priced in usage allowances, expressed as multiples of the Pro plan. The Help Center gives the Standard seat "1.25x the Pro plan's per-session usage allowance" and the Premium seat "6.25x the Pro plan's per-session usage allowance", and the pricing page describes Premium as offering "5x more usage than standard seats". A weekly limit also applies across all models, resetting at a fixed time assigned to each account, and the weekly multiple is not published. Limits belong to each member and cannot be pooled or transferred between seats.


........


Team seat

Annual billing

Monthly billing

Session allowance vs Pro

Claude Fable access

Standard

$20 per month

$25 per month

1.25x

Not listed for Standard in the pricing table

Premium

$100 per month

$125 per month

6.25x (5x a Standard seat)

50% of weekly limits on Premium seats

Enterprise seat (for reference)

$20 per month

Not offered

No seat-level limits; pay per token

Available; usage billed at Fable API rates


........


When a Team member hits a limit, the organization can let work continue through usage credits. The Help Center article on usage credits, which covers Team and seat-based Enterprise plans, says owners can pre-purchase credits, that they are billed at standard API rates, expire 12 months after purchase and are non-refundable, and that an organization-wide spend limit applies. Individual spend limits are "not available to every Team organization", according to the same article. Team also includes up to $500 a month in pooled API credits, which Enterprise does not receive.


This is the practical trade-off. A Team seat gives a predictable bill with a hard ceiling on what each person can do before credits are needed. An Enterprise seat removes the ceiling and replaces it with a meter, which suits uneven usage across a large organization and penalizes nobody for light use, but requires spend limits to keep heavy users from driving the invoice.


··········


MODELS, CONTEXT WINDOW AND PRODUCT ACCESS ON TEAM AND ENTERPRISE.


Product access is nearly identical. The pricing comparison table lists the same applications and features for both plans: Claude on web, desktop, iOS and Android, Artifacts, Claude Code, Claude Design, Claude Docs, Claude Slides, Claude Data, Claude Sheets (beta), Claude Whiteboards (beta), Claude Motion (beta), Claude Science, Claude for Microsoft 365 and Outlook, Claude in Chrome, Claude Tag, connectors, code execution and file creation, desktop extensions, enterprise search, memory, projects, research, skills, voice mode and web search. Computer use, which reached both plans in beta with an off-by-default setting, is covered in our report on Claude computer use for Team and Enterprise.


........


Capability

Claude Team

Claude Enterprise

Opus, Sonnet, Haiku

Yes ("access to all available models")

Yes

Claude Fable

Premium seats, 50% of weekly limits

Yes, billed at API rates

Context window

"Up to 1M" on pricing page; 200k in Team Help Center article

"Up to 1M", varies by model

Claude Code

Included with every seat

Included in the Enterprise seat

Cowork

Included

Included in the Enterprise seat (Help Center)

Claude Security (beta)

No

Yes

Inference hooks

No

Yes

Model training on content

None by default

None by default


........


The context window is the second documentation conflict. Both pricing tables list "Up to 1M" tokens for Team and Enterprise, with a note that it varies by model, while the Team plan Help Center article still says "200k context window". The Claude Platform pages for Opus 5.5, Sonnet 5.5, Haiku 5.5 and Fable 5.2 all list a 1M-token context window on the API.


One operational difference appears at migration rather than in the feature table. Anthropic's Team-to-Enterprise migration article says several features start switched off on Enterprise and must be re-enabled by an administrator: skills (and skill creation and sharing), code execution and file creation, Claude Design, Claude Slides, Claude Docs, standalone Claude Design and Claude in Chrome.


··········


ADMIN, IDENTITY AND SECURITY CONTROLS: WHERE CLAUDE ENTERPRISE PULLS AHEAD.


Governance is the reason most organizations move up. Anthropic's pricing table gives Team a solid baseline (SSO, domain verification, restricting organization creation, central billing, admin controls for remote and local connectors, enterprise deployment of the desktop app, usage analytics, organization-wide skills, organization instructions and spend controls) and reserves the identity-lifecycle, logging and data-governance features for Enterprise.


........


Control

Claude Team

Claude Enterprise

SSO and domain verification

Yes

Yes

JIT provisioning

Yes

Yes

SCIM provisioning

No

Yes

Domain capture

No (pricing table)

Yes

Role-based access with fine-grained permissions

No (pricing table)

Yes

Audit logs

No

Yes

Compliance API

No

Yes

Analytics API

Not listed

Yes

Custom data retention

No

Yes

Network-level access control and IP allowlisting

Not listed

Yes

Group spend limits, model choice, effort caps by role

No

Yes

Usage analytics, org skills, org instructions

Yes

Yes


........


The provisioning line is now unambiguous. Anthropic's JIT and SCIM article, dated October 15, 2026, states that "SCIM provisioning is available for Enterprise and Console organizations only" and that "SCIM is not available for Team plans". Team organizations can use just-in-time provisioning through SSO, which creates accounts on first login but does not remove them automatically when someone leaves the identity provider.


Role-based access and domain capture are less clear. The pricing table marks both as "No" for Team, while the Team plan Help Center article lists "Single-Sign-On (SSO) and Domain Capture" and "Role-based permissioning" among Team features. Team does have fixed roles (Owner, Admin and User, per the provisioning article); the fine-grained, custom role permissions described on the Enterprise card are what the table reserves for Enterprise.


The Compliance API is the feature security teams ask about most. Anthropic describes it as programmatic access to Claude usage data, "including activity logs, chat histories, and file content", filterable by user and time range, and its Claude Code page adds that it can pull Claude Code session records from the CLI and desktop app and send activity and audit logs into a SIEM. Our Claude Enterprise security guide covers the deployment side of these controls in more detail.


··········


CLAUDE ENTERPRISE COMPLIANCE: HIPAA, DATA RETENTION, US-ONLY INFERENCE AND ENCRYPTION KEYS.


Regulated organizations have no path on Team. Anthropic's HIPAA-ready Enterprise article states that Team plans and individual plans cannot enable HIPAA, while both self-serve and sales-assisted Enterprise organizations can. Only the Primary Owner can accept the Business Associate Agreement and turn the configuration on, the decision cannot be reversed from organization settings, and enabling it resets certain organization settings to defaults.


The HIPAA configuration has its own limits. Enabling it does not by itself bring Claude Code or Cowork under the BAA; Anthropic says a separate configuration is needed and is "currently only available to organizations with an Anthropic account team", and that until then Claude Code is covered only with zero data retention, which is offered to qualified accounts. API BAAs signed before December 2, 2025 do not extend to the HIPAA-ready Enterprise plan, while BAAs signed after that date can cover both.


Three more data controls are listed only for Enterprise in the Help Center: custom data retention, customer-managed encryption keys ("Provision an encryption key in your own cloud provider that your organization controls"), and US-only inference ("Keep your organization's inference within the United States"), the last of which carries the 1.1x usage multiplier noted above. Anthropic's Enterprise solutions page adds that data retention controls and OpenTelemetry monitoring are "currently available on Claude Enterprise only". No other regional data residency option is stated on the pages reviewed.


··········


CLAUDE TEAM VS CLAUDE ENTERPRISE COST FOR A 25-PERSON ORGANIZATION.


Because Team limits are expressed as multiples of Pro rather than in tokens, there is no official way to convert a Team seat into an equivalent API spend. The comparison that can be made is what an Enterprise organization would pay for defined usage profiles, against what the same headcount pays on Team seats. The profiles below are illustrative assumptions, priced at the API rates in the table above; cache-write surcharges and US-only inference are excluded unless stated.


........


Usage profile (per person per month)

Model

Input tokens

Share of input from cache

Output tokens

Enterprise usage cost

With US-only inference (1.1x)

Light chat user

Sonnet 5.5

1.5M

60%

0.15M

$2.79

$3.07

Heavy analyst

Opus 5.5

6M

70%

0.6M

$20.04

$22.04

Claude Code developer

Opus 5.5

40M

85%

1.5M

$60.80

$66.88


........


For a 25-person organization with 15 light users, 7 heavy analysts and 3 developers, Enterprise usage comes to $364.53 a month. Adding 25 Enterprise seats at $20 gives $864.53 a month, or $10,374.36 a year. On Team with annual billing, 25 Standard seats cost $500 a month ($6,000 a year); giving the 3 developers Premium seats raises it to $740 a month ($8,880 a year); and giving Premium seats to all 10 heavy users raises it to $1,300 a month ($15,600 a year). The same 22 Standard plus 3 Premium configuration on monthly Team billing costs $925 a month.


........


Configuration (25 people)

Monthly cost

Annual cost

Usage ceiling

Team: 25 Standard seats, annual

$500

$6,000

1.25x Pro per session, per seat

Team: 22 Standard + 3 Premium, annual

$740

$8,880

Premium seats 6.25x Pro

Team: 15 Standard + 10 Premium, annual

$1,300

$15,600

Premium seats 6.25x Pro

Enterprise: 25 seats + modeled usage

$864.53

$10,374.36

None beyond admin spend limits

Enterprise minimum: 20 seats, before usage

$400

$4,800

None beyond admin spend limits


........


The per-person break-even is simple. A Team Premium seat costs $100 a month on annual billing and an Enterprise seat $20, so an Enterprise user can consume up to $80 of API usage a month before costing more than a Premium seat. All three modeled profiles sit below that level, which means Enterprise's metered model is cheaper than buying Premium seats for heavy users, and more expensive than Team only where Standard-seat limits are enough. A team of 10 on Standard seats pays $2,400 a year on Team, half of Enterprise's $4,800 seat minimum before any usage.


··········


SELF-SERVE VS SALES-ASSISTED CLAUDE ENTERPRISE.


Enterprise is sold two ways, and Anthropic says both versions "include the same features and the same seat pricing". The differences are commercial.


........


Item

Self-serve Enterprise

Sales-assisted Enterprise

Where to buy

claude.ai/create/enterprise

Anthropic sales

Minimum seats

20

50 per Help Center plan article (20 on pricing page)

Usage billing

Credits purchased upfront; usage stops at zero

Billed monthly in arrears

Payment

Credit card or ACH (USD only)

Card, ACH or invoice with net terms (multi-currency)

Seat reductions

Not during the annual term

Through the account manager

AWS Marketplace, committed-spend incentives, trials

No

Yes

HIPAA-ready / BAA

Available

Available


........


The ACH path on self-serve has a delay: Anthropic says the plan is not activated until payment is received, which can take up to five business days, and auto-reload is not available with ACH. Sales-assisted contracts can also be purchased through AWS Marketplace and drawn down against an existing AWS commitment, and volume discounts for Enterprise are negotiated through sales without published rates.


··········


MOVING FROM CLAUDE TEAM TO CLAUDE ENTERPRISE: WHAT CARRIES OVER AND WHAT RESETS.


Anthropic's migration article recommends upgrading the existing Team organization instead of creating a new one, because the upgrade keeps chat history, projects, shared content, memberships and roles, while a new organization starts from scratch. The self-serve route runs from claude.ai/upgrade through "Upgrade an existing organization", requires the 20-seat minimum, asks the owner to set per-user spend limits and a starting usage balance, and takes payment by credit card at the moment of migration. The migration is not reversible.


Several practical effects follow. Unused Team usage credits roll over to the usage-based Enterprise plan, but Enterprise organizations are not eligible for the monthly API credits that Team includes; credits already claimed stay usable until they expire. If the Team organization had a spend limit, users lose access as soon as it is reached on Enterprise, so the limit should be raised before go-live. Organizations with SSO will see existing users logged out when SSO is turned on after migration, and DNS domain verification can take 24 to 48 hours. On sales-assisted migrations, users lose access during provisioning and should log out and back in afterwards.


Teams comparing Claude Team with OpenAI's equivalent before deciding whether to move up can find seat prices, Premium-seat limits and admin features side by side in our ChatGPT Business vs Claude Team comparison.


··········


CONFLICTS AND GAPS IN ANTHROPIC'S CLAUDE TEAM AND ENTERPRISE DOCUMENTATION.


Anthropic's pages disagree in four places that affect a purchase decision. The Enterprise seat minimum is 20 on the pricing page and seats article but 50 for sales-assisted contracts in the plan article. The Team context window is "Up to 1M" on the pricing page and 200k in the Team Help Center article. Domain capture and role-based permissions are marked "No" for Team on the pricing table but listed as Team features in the Team Help Center article. And the pricing table lists network-level access control and IP allowlisting on the Enterprise card without a corresponding row in the comparison table.


Several figures are not published at all: the weekly usage multiple for Team Standard and Premium seats, a maximum seat count for Enterprise, Enterprise volume discount rates, and token equivalents for Team's Pro-based multiples. The Claude Platform overview gives Fable 5.2's cache-read price only as a percentage of input. Buyers who need any of these numbers for procurement need them in writing from Anthropic.


··········


CLAUDE TEAM OR CLAUDE ENTERPRISE: WHICH PLAN FITS WHICH ORGANIZATION.


........


Organization or workload

Recommended plan

Reason

Fewer than 20 people

Claude Team

Enterprise requires at least 20 seats; Team starts at 2

20 to 150 people, mostly light chat use

Claude Team (Standard seats)

$20 per seat with usage included beats seat plus metered usage

Heavy Claude Code or Opus users

Claude Enterprise or Team Premium

Enterprise is cheaper than Premium while usage stays under $80 per person per month

Need SCIM, audit logs or Compliance API

Claude Enterprise

Marked unavailable on Team

Healthcare data under HIPAA

Claude Enterprise

Team cannot enable the HIPAA-ready configuration

Data residency or own encryption keys

Claude Enterprise

US-only inference and CMEK are Enterprise features

Need invoicing with net terms

Sales-assisted Enterprise

Team and self-serve Enterprise accept card (and ACH on Enterprise) only

More than 150 users

Claude Enterprise

Team supports up to 150 seats


........


Claude Team is the right plan for organizations of up to 150 people whose usage is predictable and whose security needs end at SSO and connector controls: a $20 Standard seat with usage included is the lowest per-person cost Anthropic offers for business use, and Premium seats cover the heaviest users at a fixed $100. Claude Enterprise is the right plan once identity lifecycle management, logging, retention, HIPAA or US-only processing are requirements, or once an organization has 20 or more people whose usage varies widely, because the metered model charges light users close to nothing and stays cheaper than Premium seats for heavy users consuming less than $80 of API usage a month.


··········


FOLLOW US FOR MORE.


·····


DATA STUDIOS


·····


datastudios.org


bottom of page