What is Streya?

Streya exists to simplify data analysis for business people. Its goal is to let a decision-maker — an executive, an operator, a team lead — ask a business question and get a business answer, without writing SQL, building a dashboard, or waiting on someone who can.

You ask in plain language — “Which accounts are declining this quarter, and why?” — and the Streya agent runs the analysis against your company’s data and answers with numbers, tables, charts, and a plain-language explanation of what it found.

The agent is built for business questions, not technical ones. It’s designed to handle the way decision-makers actually think — goals, hypotheses, “why” questions, trade-offs — and to come back with answers framed in those terms: what changed, what’s driving it, and what’s worth looking at next.

You work with Streya in two places:

Conversations — for asking and analyzing. Ad-hoc questions, follow-ups, digging into a “why”.
Dashboards — for tracking and monitoring. The KPIs you want to watch over time, assembled and maintained by prompting the agent rather than by hand.

In both, the agent presents results as three types of visuals: cards (a KPI with comparisons), charts, and tables.

How it works

Your warehouse  →  Semantic model  →  Streya agent  →  Conversations & dashboards
(BigQuery,         (cubes, views,      (plans queries,    (answers as cards,
 Snowflake)         descriptions)       runs analysis)     charts, and tables)

Connect your data. Streya connects to your data warehouse with read-only credentials. See Data connections.
Model what matters. A semantic model defines your metrics, dimensions, and business rules — what “revenue” means, how tables join, which conventions apply. Semantic layers are a well-established practice in analytics; Streya uses one so the agent works from your team’s definitions rather than inferring them from raw tables. See Semantic models.
Ask questions. Anyone on your team opens a conversation and asks. The agent picks the right dataset, runs the queries, and explains its reasoning. You can always ask how a number was calculated.
Track what matters. Describe the KPIs you want to monitor, and the agent assembles a dashboard of cards, charts, and tables you can keep refining by asking.

Who it’s for

Decision-makers — the primary user. People who consume numbers to make decisions: executives, ops leaders, sales managers, planners. They ask questions directly instead of filing a request and waiting.
The data-fluent person on the team — typically a BI lead. They work with Streya on the semantic model and the business context, so the rest of the team can self-serve. Once the model is in place, they stop being the bottleneck for ad-hoc questions.

What Streya is not

It’s not a tool for data analysts. Analysts already have SQL, notebooks, and BI suites, and those remain the right tools for their work. Streya is for the people those analysts serve.
It’s not a copy of your data. Streya queries your warehouse live with read-only access; it doesn’t ingest or warehouse your data itself.
It’s not a general-purpose chatbot. The agent only answers from your connected data and your model’s definitions. If the data can’t support an answer, it says so.

Key concepts

Concept	What it is
Workspace	Your team’s home in Streya: its data connections, semantic model, conversations, and dashboards.
Connection	A read-only link to a data warehouse (BigQuery or Snowflake).
Cube	A building block of the semantic model: one dataset’s dimensions, measures, and joins.
View	A curated, user-facing dataset combining one or more cubes. Views are what the agent analyzes.
Conversation	A chat with the Streya agent about your data.
Dashboard	A set of visuals tracking your KPIs, assembled and edited by prompting the agent.
Visual	A card (KPI), chart, or table produced by the agent from a query.

Next steps

Quickstart: ask your first question — get your first answer in five minutes.
Data connections — what to prepare to connect your warehouse.