Question 1

We already use Sigma. What does Orion add?

Accepted Answer

Orion adds the investigation Sigma leaves to your team: the recurring 'why did this move' questions stop turning into new Sigma workbooks. Sigma is a place to build on live warehouse data: spreadsheets, dashboards, and AI apps, with Sigma Assistant for natural-language queries. Orion is the analyst on top of the warehouses you already run. It watches your metrics, investigates why they moved, and delivers a written answer to Slack or email before anyone opens a workbook. Sigma is the workbench your team builds analysis in. Orion brings the finding to the people who need it.

Question 2

We build analyses in Sigma. Isn't Sigma Assistant the same as Orion?

Accepted Answer

No. Sigma Assistant, formerly Ask Sigma, is a conversational way to query your data inside Sigma. You ask a question, and it returns charts and insights you then explore in the workbook. You ask, it answers, and the analysis stays in Sigma. Orion flips what your team builds. Instead of building each analysis, you build the context once: the semantic layer Orion reads and a cited Knowledge Base the whole org reuses. Orion builds the analyses from that shared context, writes up what it found and why, and delivers it, proactively rather than waiting to be asked. Sigma Assistant helps someone in Sigma get an answer faster. Orion is the analyst that catches the change no one asked about and explains it, wherever your data lives.

Question 3

Sigma has scheduled alerts. Isn't that proactive?

Accepted Answer

Scheduled, yes. Self-directed, no. Sigma has conditional exports and alerts that run threshold and formula-defined anomaly checks on a schedule, and deliver to email, Slack, webhooks, Teams, and more. Sigma Agents run scheduled monitoring and actions. All of that is genuinely useful. The difference is that the setup is author-defined. Someone picks the metric and configures the check that decides when an alert fires. It catches the things you already knew to watch. Orion auto-detects the significant changes you did not think to define, investigates the root cause, and delivers the write-up. Sigma alerts on the conditions you set. Orion also catches the ones you did not, which is where the problems a stretched team misses actually live.

Question 4

We've invested in LookML and dbt. Does Sigma reuse them?

Accepted Answer

Partly, and less than Orion. Sigma builds its own semantic layer through Sigma data models. Its dbt integration shows dbt metadata inside Sigma, from dbt Core or dbt Cloud, but the dbt Semantic Layer connection is dbt Cloud only. Its docs show no LookML import. So your existing logic is either rebuilt in Sigma or partly connected through dbt. Orion instead reads the LookML and dbt you already trust in place, across both dbt Cloud and dbt Core. That is the logic behind every answer. A cited Knowledge Base adds the rest. Nothing to rebuild, and the semantic layer you already maintain is the one Orion uses. And if you have neither, Orion builds a semantic layer from your warehouse metadata and Knowledge Base, so there is still nothing to rebuild.

Question 5

How do the numbers stay trustworthy, and does Orion respect access controls?

Accepted Answer

Orion and Sigma both ground answers in governed logic and honor permissions. Sigma applies column- and row-level security on its data models, and grounds Sigma Assistant in admin-configured sources. Orion grounds every answer in the business logic you already trust, your LookML and dbt definitions plus a cited Knowledge Base. Every analysis is captured in a notebook: the instructions, the logic, and the code. You can see exactly how each answer was produced, and re-run it. Orion connects read-only, through a service account you scope to the data it may analyze, with optional per-user OAuth on BigQuery. Roles on two levels, tenant and group, scope who can see which projects and data. The difference is whose definitions the AI uses: Orion uses the ones you have already standardized, instead of a model you rebuild in the tool.

Question 6

Can our whole team work with Orion at the same time, or is it one person per chat?

Accepted Answer

Rooms lets several people work one question with Orion at the same time, in a single conversation. Each message carries its author, and each participant keeps their own access level, so finance can watch the investigation unfold without anyone widening permissions. Co-editing a workbook shares the document. A Room shares the analysis: one conversation your data team and your business have together, with Orion in it.

Question 7

Can we embed Orion in our own product?

Accepted Answer

Yes. Embedded Orion is a white-label analyst for your customers, presented under your brand inside your product. Each customer account is isolated: they see only their own data, business context, and permissions, and nothing passes between accounts. Your customers sign in to your product, and Orion inherits who they are and what they can see. There is no second login. You integrate once. Every new account is provisioned with its own branding, data connections, and access controls, so the second customer costs a fraction of the first.

Question 8

Can Orion and Sigma work together, or should we switch?

Accepted Answer

Yes. Orion connects to the same warehouses Sigma sits on, so they coexist. Teams keep Sigma as the place analysts and business users build spreadsheets, dashboards, and AI apps. They add Orion as the analyst. Orion monitors metrics across the whole stack, investigates what changed, and sends the written answer to Slack or email. Sigma handles the building. Orion handles the proactive investigation so the follow-up questions never pile up on the data team. And if you consolidate later, Orion already runs on the foundations your BI is built on. You migrate on your own timeline, with nothing left behind to maintain.

	Orion	Sigma
Proactive root-cause	Auto-detects a change, including ones nobody configured, investigates the root cause, and delivers a written narrative to Slack or email	Scheduled threshold and anomaly alerts you configure, sent to Slack, email, webhooks, and more. No root-cause write-up documented^↗
Everyone in one live conversation	Everyone posts into one thread and watches the same answer stream, each at their own access level. The Room stays the single record, and nobody can copy it into a private thread	Live Edit lets editors share one live workbook draft in real time. No shared Sigma Assistant thread is documented, and no shared thread is documented^↗
Semantic layer ingestion	Reads your LookML and dbt straight from their repos and reuses the dimensions, measures, and models you already defined to write better SQL. Your repo stays the source of truth. No semantic layer yet? Orion builds one from your metadata.	A Sigma data model your team builds. Its dbt Semantic Layer is dbt Cloud only. No LookML import documented.^↗
Analyst, not a tool to build in	Investigates, writes up the why, and delivers the answer to Slack or email, with nothing to build	Spreadsheet-style workbooks your team builds analyses and AI apps in^↗
Spreadsheet-style exploration	Delivers written analysis, dashboards, reports, and decks. Dashboards have filters, and you can click a number to investigate it. There is no spreadsheet grid to work in	A spreadsheet interface over live warehouse data that business users already know^↗
Dashboard and AI-app authoring	Generates dashboards, reports, and decks on request. You refine by describing the change, not by building by hand	Dashboards, AI apps, and write-back your team builds hands-on^↗

Orion vs Sigma

TL;DR

What is the difference between Orion and Sigma?

How do Orion and Sigma compare feature by feature?

Who builds the analysis: your team or Orion?

When should you choose Sigma instead?

Orion vs Sigma: what do buyers ask most?

More Orion comparisons