Ask with context
- Plain-English questions and read-only analytical queries
- Tables, charts and short business insights
- Assumptions explained and ambiguous wording clarified
- Source fields and visible trust scores for returned results
MEET ZENTARRA 4.5
From a business question
to repeatable reporting.
Explore governed definitions, visible result checks and reusable reporting in Zentarra, the AI analytics product developed and sold by AICowan.
AVAILABLE IN ZENTARRA 4.5
From the first question
to everyday reporting.
A managed early-access service with practical tools for exploring governed business data. Availability and configuration are agreed during onboarding.
Build on useful answers
- Saved queries and reusable templates
- Dashboard building and sharing
- CSV export of displayed results, plus chart and report exports
- Scheduled dashboard refreshes when enabled
Keep access in scope
- Workspace-specific users, permissions and data access
- Microsoft Entra ID and time-limited magic-link authentication
- Administrative and operational audit records
- Analysis blocking when critical data-quality controls require it
Trust scores reflect the configured checks, including semantic governance, executed results, field coverage and joins. They help you assess an answer alongside data-quality information. A score is not a probability of correctness or a guarantee that the result is suitable for a business decision.
DATA CONNECTIONS
Start with the right connection.
Azure SQL Database is the currently proven production and demonstration connection.
Azure SQL Database Proven connectionADDITIONAL CONNECTORS · SUBJECT TO VALIDATION
Microsoft SQL ServerPostgreSQLMySQLSnowflakeGoogle BigQueryMicrosoft Fabric WarehouseMicrosoft Fabric Lakehouse
Connector implementations are available subject to validation. Current evaluations use an online data source controlled by AICowan; customer database connections are outside the current evaluation offer.
WATCH THE PRODUCT
Look beyond
the answer.
Gary’s recorded walkthrough explores data-quality validation, trust scores, freshness indicators and analysis blocking when critical checks fail.
This recording demonstrates an earlier stage of Zentarra. The current interface and available configuration may differ. See the Product page for the current 4.5 capabilities.
Trust & data quality walkthrough
Watch the Zentarra walkthrough (opens on YouTube in a new tab)Opens on YouTube. No video player is loaded on this website.
Read the walkthrough overview →INSIDE ZENTARRA
A real question.
A result you can explore.
See Zentarra in action: return costs grouped by reason, a visual comparison, and the supporting figures together in one workspace.
Zentarra · Product screenshotView full-size image ↗
Read the screenshot as text
Zentarra V4.5 answers: “Show total return cost by return reason, combining all years.” The result has eight rows and three columns. A bar chart compares return costs; further categories are available through its horizontal scrollbar in the captured application.
| Return reason | Return cost | Returned shipments |
|---|---|---|
| Recall | 155,194.75 | 3,447 |
| Address issue | 151,620.82 | 3,429 |
| Damaged | 151,444.83 | 3,447 |
| Wrong item | 150,138.73 | 3,438 |
| Failed delivery | 148,428.20 | 3,446 |
| Customer refused | 146,778.51 | 3,432 |
| No longer required | 10,590.75 | 100 |
| Duplicate order | 9,450.16 | 108 |
The captured result displays “Trust score: 100/100 · Safe” and says it uses saved definitions and checks. This describes the captured checks, not a guarantee of correctness. It also reports “Save-ready rerun unavailable: unsupported shape”.
The interface shows a displayed-values CSV option, Export CSV and View details controls. These are part of the static screenshot, not interactive controls on this website.
A LITTLE CURIOSITY GOES A LONG WAY
What would you
ask your data?
From the big picture to the next detail, start with the questions that matter to your business.
AN EXAMPLE WITH SAMPLE DATA
REVENUE AT A GLANCE
£284k↗ 18.3% quarter on quarter
See the direction, not just the total.
Compare performance over time and use the underlying query to understand what went into the calculation.
Illustrative figures. This website does not query live business data.
YOUR NEXT QUESTION COULD CHANGE THINGS
Request a guided
Zentarra evaluation.
Request an evaluation ↗
Start with an enquiry. We’ll discuss the next steps.