Epic Cogito comes up in executive meetings in two ways: as the reason a dashboard should be quick to build, and as the reason two reports disagree. Both are really questions about governance, not software.
This guide is for CIOs, CMIOs, chief data officers and operational leaders at health systems that run Epic, and for vendors who sell to them. It explains what Cogito gives leaders, how executive dashboards should be designed, who should own metric definitions, when to build in Cogito versus an outside tool, and where analytics programs stall.
What is Epic Cogito for executives?
Joining outside data to Epic analytics?
Agnotic builds Epic integrations for health systems and vendors, including governed data feeds that line up with the metrics leaders see in Cogito.
Epic Cogito, for executives, is the analytics layer of Epic: Radar dashboards, Reporting Workbench reports and SlicerDicer self-service queries, all drawing on Epic's data stores. Leaders use it to watch operational and clinical measures inside the same system clinicians use, so a number on a dashboard can lead straight to the patients or orders behind it.
UC Davis Health describes Cogito as Epic's module for analytics, made up of Radar, Reporting Workbench, SlicerDicer, the Analytics Catalog and the Chronicles, Clarity and Caboodle data stores. For a leader, the tools matter more than the stores, because the tools decide how fast a question turns into an answer.
Which Cogito tools will leaders actually use?
Leaders mostly use Radar dashboards, with SlicerDicer for follow-up questions and Reporting Workbench for lists their teams act on. The University of Iowa Health Care describes Cogito Palette as a single experience for visualizing data from Reporting Workbench, Radar and SlicerDicer.
- Radar: UC Davis Health describes Radar as a business intelligence tool in Cogito whose dashboards hold visualizations, metrics, links to activities and reports, and links to external reports such as Tableau dashboards.
- SlicerDicer: A self-service tool for exploring patient populations, trends and line-level detail; UC Davis notes that viewing patient details and exporting data require approvals and permissions.
- Reporting Workbench: Operational reports that return lists of patients, orders or accounts, which staff can act on directly from the results.
- Analytics Catalog: The place users find the dashboards, reports and SlicerDicer sessions they have access to.
UC Davis Health also notes that every user role in Epic should have a default dashboard customized for its needs, and that Radar components can be built from SlicerDicer queries. That makes the dashboard the front door: if the default view for a role is cluttered or wrong, people go back to spreadsheets.
How should an executive dashboard be designed?
An executive dashboard should answer a short list of agreed questions, with each metric defined once, owned by a named person and linked to the detail behind it. Fewer metrics with clear owners beat a wall of charts every time.
- 1. Questions first: Write the five to eight questions the leadership team asks every week, such as how full the beds are or how long new patients wait for a visit.
- 2. One metric per question: Pick the measure that answers it, and write its definition, numerator, denominator and exclusions.
- 3. Targets and thresholds: Set what good looks like, and what level triggers a conversation.
- 4. Drill path: Link each tile to the department view and then to the patient or order list behind it.
- 5. Refresh and lag: Show when each metric last refreshed, because some update in near real time and others after overnight loads.
- 6. Start from Epic's content: UC Davis Health points to foundation Radar dashboards in Epic's Report Repository, which a team can review before building its own.
Resist adding a tile because someone asked once. Every metric on an executive view creates follow-up work for an analyst, so each one should earn its place by changing a decision.
Who should govern Cogito metrics and definitions?
A small analytics governance group, chaired by an executive sponsor and staffed by metric owners, should govern Cogito metrics and definitions. Without it, each department builds its own version of the same measure and leaders lose trust in all of them.
- Executive sponsor: Settles disputes about definitions and decides which metrics appear on leadership dashboards.
- Metric owners: Operational leaders who own each definition and answer for its accuracy, not the analysts who build it.
- Analytics team: Builds and maintains dashboards, documents logic, and tests changes against prior figures.
- Certification: A visible mark on dashboards and reports that the governance group has reviewed, so users can tell official numbers from personal ones.
- Access review: Rules for who can see patient-level detail and export it, matching the approvals the health system already requires for SlicerDicer exports.
- Change log: A record of every definition change with its date, so a jump in a trend can be traced to a rule change rather than a real shift.
Meet monthly at first, with a standing agenda: new requests, definition disputes, certification decisions and retirements. Retiring unused dashboards is as important as building new ones.
What should a CIO ask before funding a Cogito dashboard program?
A CIO should ask which decisions the dashboards will change, who owns each measure, and how much analyst capacity the program will consume after launch. A dashboard program is an ongoing service, not a one-time build, and its running cost sits mostly in people.
- Decisions: Which meetings will use the dashboard, and what will they do differently when a number moves?
- Ownership: Is there a named operational owner for every proposed metric, and have they agreed the definition in writing?
- Existing content: Has the team checked Epic's foundation dashboards and the health system's own catalog before building something new?
- Capacity: How many analyst hours per month will maintenance, fixes and change requests take once the dashboard is live?
- Training: How will managers learn to drill from a tile to the patients behind it, and who answers their questions?
- Measures of use: How will the program know whether leaders open the dashboard, and what happens to views nobody uses?
These questions also shape vendor contracts. A vendor that adds metrics to a health system's dashboards adds to the analytics team's workload, so the contract should state who maintains those views and who answers when a number looks wrong.
When should leaders build in Cogito versus an outside tool?
Build in Cogito when the measure is about Epic data and the people acting on it work in Epic. Use an outside BI tool when the measure combines Epic data with finance, staffing or supply systems, or when the audience never opens Epic.
- Clinical operations: Cogito, because clinicians and managers can act on the list without leaving Epic.
- Quality measures: Cogito where Epic content already exists, with outside validation for measures reported to regulators.
- Enterprise finance: An outside tool when cost, general ledger or payroll data sit outside Epic.
- Board reporting: An outside tool or a curated export, so the board pack is stable and versioned.
- Vendor and partner metrics: Usually outside Epic, joined to Epic extracts through agreed keys.
- Blended views: Radar can link to an external dashboard, which keeps one front door for leaders even when the detail lives elsewhere.
How do outside vendors and data fit with Cogito?
Outside vendors fit with Cogito by sending data the health system can load and govern, and by aligning their metrics with the definitions the governance group already uses. A vendor that reports different numbers from the customer's own dashboards loses the argument, even when its maths is right.
In practice that means agreeing on identifiers, definitions and timing before go-live. Patient and encounter keys must match Epic's, a measure such as readmission must use the customer's exclusions, and the vendor's refresh schedule must line up with the customer's overnight loads. Live data for apps should still come from FHIR APIs or interfaces, not from analytics stores.
Data that flows back into Epic, such as risk scores or program enrollment flags, needs a defined home in the record. Once it is stored as structured data, the analytics team can surface it in Reporting Workbench lists and dashboards like any other Epic field.
Where do Cogito programs stall?
Cogito programs stall when the tools are switched on but nobody owns the questions, the definitions or the backlog. These are the most common patterns.
- Dashboard sprawl: Hundreds of personal dashboards, few certified ones, and no way to tell them apart.
- Duplicate definitions: Two departments report length of stay with different exclusions and both are right by their own rules.
- Analyst bottleneck: Every request goes to a small team, and leaders wait weeks for small changes.
- Unused defaults: Role dashboards are never tailored, so staff ignore them.
- Export creep: Patient-level exports multiply without review, outside the controls the health system set.
- No retirement: Old reports keep running and confuse new users.
Most of these trace back to the same gap: the health system bought and configured the tools but never staffed the governance work around them. Fixing that rarely needs new software. It needs a sponsor who protects analyst time, metric owners who show up, and a short list of certified views that leaders agree to use in their own meetings.
A quick health check helps before any new build. Count certified dashboards against personal ones, list the metrics with more than one definition, and ask each executive which view they opened last week. The answers usually point to the first three fixes.
How Agnotic can help
Agnotic Technologies helps health systems and the vendors that serve them make outside data work alongside Epic analytics. For the client, Agnotic builds governed data feeds, matching keys and definitions, extract pipelines from Epic data stores, and integrations that write structured results back into Epic.
Agnotic also builds the dashboards and data models that sit outside Epic when finance, staffing or partner data is involved, so leaders get one consistent set of numbers wherever they look.
Align your data with Epic analytics
See how Agnotic approaches Epic integration work, or read further in the EHR integration hub.
Related reading on Agnotic
Continue with Epic Caboodle, Epic Clarity extracts, EHR data analytics, healthcare KPIs and EHR integration methods.
