Omni

Overview

Omni's pitch is aimed at a specific and familiar failure: a governed semantic model that nobody can move fast enough inside, or fast ad hoc SQL that nobody can trust afterwards. Its answer is to refuse the choice — analysts work in SQL, and Omni builds the shared model out of what they do, so the model accumulates from real analysis rather than being specified up front.

That is the distinguishing claim in this category. Lightdash inherits its model from dbt, Sigma governs a spreadsheet surface, and Omni grows a model as a by-product of use, which it presents as giving reusability, governance and query performance without a modelling project first.

Beyond internal reporting it targets customer-facing data products — reports embedded in someone else's product — and makes a point of the cost of that, arguing teams should not have to pay enterprise prices or burn engineering time to build and iterate on external-facing reports.

Security is described concretely rather than by adjective: row-level and column-level permissions, content controls, user attributes and SSO.

For programmatic use there are REST APIs, a CLI, and agent skills for bringing Omni workflows into an agent.

Pricing is not published, and unlike its competitors there is not even a pricing page: /pricing returns a 404. Administration documentation covers billing, so the model exists — it simply is not public.

Key features and capabilities

The same headings are used for every bi & dashboards entry, so two tools can be read side by side.

Data modelling and semantic layer

• A shared data model built progressively from the analysis people actually perform, not specified up front
• SQL and the model coexist, rather than the model being a gate in front of SQL
• Metrics defined once in the shared model and reused anywhere

Charts, dashboards and interactivity

• Out-of-the-box visualisations plus customisable components including links, text and images
• Aimed at both internal dashboards and polished customer-facing reports

Calculations, AI and natural language

• Immediate exploration with instant query results as the stated goal
• Agent skills bring Omni workflows into an agent's reach

Publishing, embedding and collaboration

• Internal and external-facing reports from the same platform
• Customer-facing data products treated as a primary use case, not an add-on

Security, lineage and certification

• Row-level and column-level permissions
• Content controls and user attributes
• SSO authentication
• Standardised metric definitions in the shared model as the governance mechanism

Integrations

• REST APIs for programmatic interaction
• A CLI wrapping those APIs for terminal use
• Agent skills for agent-facing workflows

Where it runs and what it costs

• Cloud service with a free trial
• Administration covers users, billing and authentication in-product
• No published pricing; there is no pricing page on the site

Pricing

Price on request

Not published, and not even partially — the site has no pricing page and /pricing returns a 404, so tiers, seat model and list prices are all unavailable publicly. A free trial is offered. Worth noting against its own marketing, which argues teams should not pay enterprise prices for customer-facing reporting.

Vendor pricing page →

Demos and videos

• Omni documentation Tutorial · docs.omni.co (the vendor)

About Omni

Omni Analytics builds a BI platform combining a shared data model with SQL, aimed both at internal analytics and at customer-facing data products. Its public case studies name Synthesia, Currys and dbt Labs. Documentation is open and covers administration, modelling, REST APIs, a CLI and agent skills. It publishes no pricing page at all — /pricing returns a 404 — so licensing and cost can only be established through the company, despite its marketing arguing explicitly about not paying enterprise prices.

omni.co