dbt Cloud Pricing and dbt Cloud Enterprise Pricing per Seat in 2026

dbt Cloud pricing has four plans. Developer is free for one seat, Starter is $100 per developer seat a month for up to five seats, and Enterprise and Enterprise+ are quoted by sales. The only public Enterprise-class price is dbt Labs' own AWS Marketplace listing: $48,000 a year for 10 developer licenses, which is $400 per seat a month, four times the Starter rate. Usage sits on top: successful models built, Semantic Layer queried metrics, dbt State at $0.094 per daily active target table and dbt Wizard AI credits.

Search for dbt Cloud pricing and page one is third-party guides: Hevo, Paradime, b-eye, a couple of pricing directories. They disagree with each other about Enterprise by an order of magnitude, from about $4,800 a seat to "$90,000 to $400,000 a year", and none of them cites a primary source for the number. dbt Labs' own pricing page prints two prices and two "Contact sales" buttons. So we read the three primary sources that exist: the pricing page, dbt Labs' service consumption table, and the dbt Platform listing that dbt Labs sells through AWS Marketplace. Everything below was read on 9 October 2026.

dbt Cloud pricing at a glance

From getdbt.com/pricing and its plan comparison grid, read 9 October 2026. US list prices.

The jump that matters is between Starter and Enterprise. Starter stops at five developer seats and one project, so the sixth person who writes models forces the conversation with sales, and so does the first request for SSO or column-level lineage.

How much dbt Cloud Enterprise costs

dbt Labs does not print an Enterprise price on its website. It does sell the dbt Platform through AWS Marketplace, and a transactable listing has to carry a real number. The listing has one dimension, described as "dbt Platform Plan: 10 Develop Licenses", priced at $48,000.00 for 12 months. No 24 or 36-month term is offered there.

Divide it out and you get $4,800 per developer seat a year, or $400 a month. That is exactly four times Starter's $100. It is also the only Enterprise-class figure with a primary source behind it, and it lines up with the low end of the third-party estimates. We found no primary source for the high-end figures you will see quoted ("$90,000 and up"). Larger seat counts, extra projects and consumption commits are what would push a real quote in that direction.

Treat $48,000 as list for ten seats, not a quote. Large accounts negotiate, and Enterprise contracts are commonly structured around an annual commit that consumption draws down. dbt's billing docs say Enterprise admins see "how much you have drawn down from your annual commit and how much remains", which tells you the commit is part of the deal.

How the dbt Cloud bill is built

Seats

dbt bills "primarily on the amount of licenses purchased". There are four license types: Developer for day-to-day work, IT for account management such as configuring the git integration, Read-Only for viewing documents and reports, and Analyst, which dbt now marks as "not available for new purchase". Anyone who writes, tests or deploys models needs a Developer seat. If your plan is to give fifty analysts access to the catalog, ask in writing how Read-Only licenses are priced on your contract.

Successful models built

This is the usage meter people misjudge. A successful model built is any model that builds in a deployment environment through dbt's orchestration, and that includes scheduled jobs, CI builds triggered by pull requests and runs started through the API. Models count "even if the overall run later fails": a job that dies after 51 of 100 models still bills 51.

The math climbs fast. A 300-model project built once a day is about 9,000 models a month, inside Starter's 15,000. Build it every four hours and you are at about 54,000, well past Starter and halfway to Enterprise's 100,000, before a single CI run. Neither the pricing page nor the billing docs print an overage rate for Starter, so get it in your order form before you schedule hourly jobs.

Semantic Layer queried metrics

Every successful request to render or run SQL through the Semantic Layer API counts as at least one queried metric, even if no rows come back, and each metric in a multi-metric query counts separately. Compiling a metric counts the same as running it. Starter includes 5,000 a month and Enterprise 20,000. A busy BI tool can spend that faster than you would expect. If the people asking for numbers would rather type a question than learn an API, an AI data analyst that turns plain-English questions into SQL is the other way to serve them.

dbt State and dbt Wizard

Two newer add-ons are billed by consumption with "no seats, no minimums". dbt State skips, clones or reuses models that have not changed, and dbt Labs' service consumption table, effective 1 July 2026, bills it at a list rate of $0.094 per daily active target table. A target table includes every distinct test, so a model with a not_null and a unique test on one column is three target tables. Reused models stop counting as successful models built, which is the trade: fewer model builds, a new meter. dbt Wizard is the AI agent, billed per token by model, with a $100 trial credit on Developer and Starter and $100 or $200 a month of included credit on Enterprise and Enterprise+.

What a real team pays for dbt Cloud

The cliff is the point. Going from five developers to six moves a team from $6,000 a year to a sales-led contract whose only public anchor is $48,000. Budget for that step a quarter before you hire the sixth engineer, not the week the invitation fails.

Is dbt Cloud Enterprise worth it for column-level lineage?

Often it is the feature that triggers the upgrade. On the pricing grid, column-level lineage in dbt Catalog is an Enterprise feature; Starter gets model-level lineage only. Model performance and query history sit on the higher plans too. For a team that is otherwise happy on Starter, that makes column lineage a $42,000-a-year question at list: the difference between five Starter seats and the ten-license platform plan.

If the real need is to see which columns and dashboards a change reaches, there are cheaper ways to get it. Our breakdown of dbt column-level lineage covers every option across dbt Cloud, dbt Core and Fusion. DataTrail builds column lineage from your dbt artifacts and the warehouse's own query history with a read-only role, and carries it through to Tableau, Looker and Power BI, from $708 a year on Starter or $2,148 on Team, with Enterprise above that. It does not replace dbt Cloud's IDE, scheduler or Semantic Layer, and it is not trying to. It answers the "what breaks if I change this" question that usually starts the upgrade conversation, and our comparison of dbt observability tools prices the monitoring options that sit beside it.

dbt Cloud pricing questions

How much does dbt Cloud cost?

dbt Cloud is free for one developer on the Developer plan and $100 per developer seat a month on Starter, which allows up to five seats and 15,000 successful models built a month. Enterprise and Enterprise+ are quoted by sales. The only public Enterprise-class price is dbt Labs' AWS Marketplace listing at $48,000 a year for 10 developer licenses, read 9 October 2026.

How much does dbt Cloud Enterprise cost?

dbt Labs does not publish an Enterprise price on its website. Its AWS Marketplace listing sells the dbt Platform at $48,000 for 12 months for 10 developer licenses, which works out to $400 per seat a month. Larger seat counts, more projects and consumption commits raise that, and negotiated discounts lower it.

Is dbt Cloud free?

Only for one person. The Developer plan is free with one developer seat, 3,000 successful models built a month and one project. The open source dbt Core is free to license but you run, schedule and host it yourself, which costs engineer time and infrastructure instead of seats.

What counts as a successful model built in dbt Cloud?

Any model that builds successfully in a deployment environment through dbt's orchestration, including scheduled jobs, CI builds from pull requests and API-triggered runs. Models count even if the run later fails. Models that dbt State skips or clones do not count as models built; they are billed separately as daily active target tables.

Does dbt Cloud Starter include column-level lineage?

No. Starter includes lineage in dbt Catalog at the model level. Column-level lineage is listed as an Enterprise and Enterprise+ feature on dbt's pricing page, read 9 October 2026. Teams that need it on Starter typically add a separate lineage tool rather than upgrade every seat.

Where these numbers came from

Plan prices, seat counts, model and metric allowances and the feature grid come from getdbt.com/pricing. License types and the definitions of a successful model built and a queried metric come from dbt's "How does dbt pricing work?" documentation. The dbt State rate and the target table definition come from the dbt Labs Service Consumption Table, effective 1 July 2026. The Enterprise-class price is the dbt Platform listing by dbt Labs on AWS Marketplace, read from its rate card on 9 October 2026. All are US list prices before private offers or discounts. If you are mapping a dbt project before a renewal, the dbt lineage view shows which models and dashboards depend on what, so you can size the seats and schedules you actually need.