Analysts in control. Governed insights at scale.

Aria is an AI insights platform from Enlighten. Analysts define the methods, and AI agents execute them to produce insights for the business. It is built for analysts and analytics leaders who want their expertise to reach further, and for knowledge workers who need clear, reviewed answers grounded in their organization's data and business definitions.

Yes. Recurring analysis is a core use for Aria. An analyst defines the method once, including the data, metric definitions, prompts and code it should use, and agents execute that method each time it runs. Because the method and business definitions stay consistent, each run is comparable to the last, and findings still pass through verification and human review before they are shared.

BI tools are strong at dashboards that show what happened, and chatbots answer one-off questions. Aria executes analyst-defined methods to produce findings, then checks them through its 4D Evaluation and human review before they are shared. Those governed findings and their evidence are kept in the Insights Library, and Chat answers follow-up questions grounded in those insights.

Analysts define the method, and agents execute it. In the Insights Factory, analysts build methods in the visual Composer or work from their preferred coding tools through the Aria CLI. Either way, the method draws on the Intelligence Foundation: your data, Memory, and AI Instructions such as prompts, code and analytical guidance. Agents then run that work, and findings go through human review before they are shared.

Every finding passes through Aria's 4D Evaluation before it is shared. The four dimensions are Runtime Evaluation, Claims Extraction, Claims Verification and Observation, followed by Human Review, so people stay involved in what is shared. Governed findings are kept with their supporting evidence in the Insights Library, so your team can see where a conclusion came from and revisit it later.

Data alone does not tell an agent what your numbers mean. Aria pairs your data with Memory, which acts as a semantic layer of shared business meaning: metric definitions, approved rules, context and prior work. When an analysis calculates a metric or applies a rule, it draws on the definitions your organization has agreed on, so results stay consistent and comparable across teams and locations.

Memory is Aria's business-meaning and knowledge layer. It holds your metrics, approved rules, business context and prior work, so analysis reflects how your organization defines and understands its business. The Dream Cycle is opt-in catalog maintenance. It surfaces issues in the catalog for people on your team to resolve, so decisions about shared business meaning stay in human hands.

Aria shares insights through the Insights Experience: bespoke, highly interactive reports rather than static slides or dashboards. Readers can use Chat to ask follow-up questions, with answers grounded in the insights behind the report. DoubleClicks open the supporting detail behind a finding, so readers can explore the evidence for themselves instead of taking a summary at face value.

Yes. Aria supports scheduled, parameterized distribution, so one analyst-defined method can run for each store, restaurant, property or region and reach hundreds or thousands of locations. Each location receives findings specific to its own data, built on the same metric definitions and checks as every other location. That keeps local insights relevant while the method behind them stays consistent across the organization.

Aria is available as Enlighten-managed SaaS or as Bring Your Own Cloud on Google Cloud or AWS, so you can choose the option that fits your security and infrastructure requirements. Administrators control which AI models and tools are available, who has access, how usage is managed, and the audit policies that apply. These controls let your organization set clear boundaries for how Aria is used.

Aria is a strong fit for multi-location businesses such as quick-service restaurants, hospitality and retail, where the same analysis often needs to run for many stores or properties. Recurring performance analysis and location-level reporting are natural starting points. It also suits analytics teams that want analysts to define methods once and deliver reviewed insights to knowledge workers across the business.