Health Cloud CRM Analytics Settings
Health Cloud CRM Analytics Settings is a Setup page for configuring the CRM Analytics app dedicated to Salesforce Health Cloud.
Definition
Health Cloud CRM Analytics Settings is a Setup page for configuring the CRM Analytics app dedicated to Salesforce Health Cloud. The settings page lets healthcare organizations deploy pre-built dashboards and datasets that read from Health Cloud's clinical data model: patients, care plans, care team members, clinical encounters, conditions, medications, and assessments. The result is a starting view of population health, care coordination effectiveness, and outcome trends without building dashboards from scratch.
The page sits in Setup under CRM Analytics, with a Health Cloud subsection that bundles the wizard that deploys the analytics app plus the configuration options for dataset refresh schedules, row-level security predicates, and tier-based filtering. Most healthcare orgs running Health Cloud also license CRM Analytics, and the Health Cloud Analytics Settings is the connecting tissue between the operational system and the analytical surface that clinical and operational leadership relies on.
What the Health Cloud Analytics deployment includes
Pre-built dashboards
The settings page deploys around six to eight dashboards covering the core healthcare workflows. Population Health shows patient cohort breakdowns by condition, age band, payer, and risk score. Care Coordination tracks care plan completion rates, gaps in care, and assigned care team workload. Clinical Outcomes visualizes outcome metrics like blood pressure control rates for hypertensive patients, hemoglobin A1c control for diabetics, and screening completion rates against benchmark targets. Care Team Performance measures clinician and care manager activity, response times, and patient satisfaction scores. Each dashboard supports filters by clinic, payer, and care team, with drill-through to the underlying patient list when an outlier needs investigation.
Datasets and dataflows
The deployment creates several datasets specific to healthcare. Patient Population combines Account, Contact, Patient (Person Account in Health Cloud), HealthCondition, Medication, and CarePlan rows. Care Encounters joins clinical events with assigned care team members. Outcomes Tracking pulls observation-style data (lab values, vitals) and matches them against care plan goals. Each dataset is refreshed via a dataflow defined in JSON, visible and editable in the Analytics Data Manager. The default refresh schedule is nightly during the org's quietest hours, with the option to switch to hourly refreshes for clinical dashboards that need closer-to-real-time data.
Row-level security for HIPAA compliance
Health Cloud Analytics enforces row-level security through dataset predicates that filter records based on the running user's role and assigned patient cohort. A care manager assigned to specific patients sees only those patients in their dashboards. A clinic director sees only patients of their clinic. Population health analysts see aggregated data only, with patient-level drill-through restricted unless they have explicit permission. The predicates are defined in the dataflow JSON during the wizard run, but ongoing maintenance happens through the Analytics Data Manager. HIPAA-compliant deployment requires careful predicate design and quarterly verification that the predicates still match the org's role hierarchy.
Patient-cohort definitions and segmentation
The Health Cloud Analytics app supports patient cohort definitions: saved groupings of patients matching specific clinical criteria (all diabetics, all post-discharge patients in the first 30 days, all patients overdue for annual wellness visits). Cohorts are defined inside the analytics app and become filter values across every dashboard. The cohort definitions are usually built by a clinical analyst who knows the relevant clinical guidelines (American Diabetes Association A1c targets, CDC adult immunization schedules), and they often drive the org's quality improvement initiatives. Each cohort can be refreshed alongside the dataset refresh, so new patients flowing into Health Cloud are automatically included.
Integration with Einstein Discovery
Health Cloud Analytics integrates with Einstein Discovery for predictive use cases: identifying patients at risk of readmission within 30 days, predicting which patients will miss appointments, scoring lead probability of converting to a managed-care contract. The integration appears as predicted-value columns inside the dashboard tables, generated by Discovery stories trained on historical Health Cloud data. The setup of each Discovery story is a separate effort with its own data prep and model validation, but once it is running, the predictions flow into the dashboards automatically. Healthcare organizations that adopt predictive analytics typically do so after their descriptive analytics dashboards are stable.
Customization paths after deployment
The deployed dashboards and datasets are editable like any CRM Analytics asset. New widgets, filters, and pages can be added through the Dashboard Builder. New columns can be added to the datasets by editing the dataflow JSON. Best practice is to clone the wizard-deployed dashboards before customizing so a future re-run of the wizard (after a Health Cloud upgrade introduces new fields) does not overwrite the customizations. The clones become the production-facing assets while the originals remain as a baseline reference. This pattern is the same as in other industry analytics wizards (Financial Services, Manufacturing) but matters more in healthcare because data model changes are constant.
Limits, scale, and refresh windows
Healthcare orgs sometimes run into CRM Analytics scale limits. Population dashboards spanning hundreds of thousands of patients can produce dataflows that take an hour or more to refresh. The settings page exposes the dataflow run history and lets admins see which datasets are bottlenecks. For very large deployments, the right path is to split the dataset by line of business (Commercial Insurance versus Medicare versus Medicaid) into separate datasets refreshed on staggered schedules. Big organizations also use the recipe-builder approach (ELT inside CRM Analytics) instead of the legacy dataflow JSON, because recipes scale better and offer richer transformations.
Pitfalls specific to healthcare analytics deployments
Three pitfalls recur in Health Cloud Analytics rollouts. First, predicate misalignment after a care team reorganization: when a clinical leader is reassigned to a different care team, the row-level security predicate continues to reference the old assignment until the dataflow refreshes, which can leave that user looking at the wrong patient cohort until the next refresh runs. Schedule predicate verification immediately after any major care team change. Second, dataset drift due to upstream schema changes: Health Cloud's data model evolves with each release, and new fields the analytics dataset does not pick up cause "we know this data exists but cannot see it" complaints from clinical leaders. Plan a quarterly dataflow review against the Health Cloud release notes. Third, performance issues with longitudinal cohort analyses: dashboards that try to show three years of trend data for hundreds of thousands of patients often hit dataflow limits. Pre-aggregate the historical data into a separate dataset rather than recomputing the trend on every refresh. None of these pitfalls is exotic, all of them appear in real healthcare orgs, and Health Cloud Analytics Settings is the page where the configuration to address them lives.
Deploy and configure Health Cloud Analytics
Deploying Health Cloud Analytics is a wizard-driven process layered on top of a working Health Cloud configuration. The walkthrough below covers the standard go-live sequence from prerequisites through dashboard customization with clinical stakeholders.
- Confirm Health Cloud and CRM Analytics prerequisites
Verify that the org has Health Cloud licensed and configured with at least one Patient (Person Account) record, one Care Plan, and one HealthCondition record. Confirm that CRM Analytics Plus is licensed and that the running admin user has Manage CRM Analytics permission. Without these prerequisites, the wizard runs but the dashboards return zero results, which causes a confusing first demo to clinical stakeholders.
- Run the Health Cloud Analytics wizard
From Setup, search for Health Cloud CRM Analytics Settings and select the page. Click Start to launch the wizard. The wizard prompts for an app name, an initial set of filters (by line of business or geographic region), and a deployment confirmation. Click Deploy. The deployment runs for fifteen to forty-five minutes, depending on data volume. Salesforce sends an email when the deployment completes, after which the dashboards become visible in the Analytics Studio under the new Health Cloud Analytics app.
- Configure row-level security predicates
Open the Analytics Data Manager, find the Patient Population dataset, and inspect the security predicate. Compare it to the org's actual role hierarchy and Health Cloud care team assignments. Adjust the predicate JSON to match the org's clinical access controls. Test by logging in as a sample care manager and confirming they see only their assigned patients in the dashboards. Repeat for every dataset that holds patient-level data. This step is the gating step for HIPAA-compliant rollout and should not be skipped.
- Customize, validate, and roll out
Clone every wizard-deployed dashboard so customizations do not get overwritten by future wizard re-runs. Add cohort filters relevant to the org's quality initiatives. Add organization-specific KPIs the standard dashboards do not include. Walk the clinical leadership team through each dashboard, capture feedback, iterate. Schedule a soft launch for two weeks with a small group of clinical users, capture feedback, fix issues, then announce broadly. Communicate the dashboards' refresh schedule clearly so users do not expect intraday data when the refresh is nightly.
- The wizard does not validate that source data is present. Running against an empty Health Cloud setup produces dashboards that load successfully but show no patients anywhere.
- Re-running the wizard overwrites the deployed dashboards. Always clone before customizing so a future wizard run does not erase the changes.
- Row-level security predicates that look correct can still expose patient data if the care team assignments are stale. Refresh the test scenario after any role hierarchy change.
- Default dataflow schedule is nightly. Clinical stakeholders expecting real-time visibility into newly admitted patients need to be set straight before the first live demo.
- The Health Cloud Analytics app requires CRM Analytics Plus. Orgs on the base CRM Analytics SKU see the settings page but cannot complete the deployment.
Trust & references
Cross-checked against the following references.
- CRM Analytics for Health CloudSalesforce Help
- Health Cloud OverviewSalesforce Help
Straight from the source - Salesforce's reference material on Health Cloud CRM Analytics Settings.
- Health Cloud AnalyticsSalesforce Help
- CRM Analytics SecuritySalesforce Help
About the Author
Dipojjal Chakrabarti is a B2C Solution Architect with 29 Salesforce certifications and over 13 years in the Salesforce ecosystem. He runs salesforcedictionary.com to help admins, developers, architects, and cert/interview candidates sharpen their fundamentals. More about Dipojjal.
Test your knowledge
Q1. What business function does Health Cloud CRM Analytics Settings primarily support?
Q2. What customer experience metric would Health Cloud CRM Analytics Settings help improve?
Q3. Which Salesforce Cloud includes Health Cloud CRM Analytics Settings as a key feature?
Discussion
Loading discussion…