EV Healthcare: Becoming A Dashboard Jedi

Most compliance teams are not short on data. They are short on information they can act on. In this healthcare edition of The Ethicsverse, Caryn Weferling, Director of Compliance at ScionHealth, and William Chiles, Regional Compliance Director at ScionHealth, joined host Nick Gallo, Chief Servant & Co-CEO of Ethico, to share how they turned scattered spreadsheets and pivot tables into a library of focused, self-refreshing dashboards.

William came to compliance from an IT background and began by mapping where the department's data lived, extracting it, and building simple visuals that stakeholders could click, filter, and drill into in real time. Caryn, who describes herself as having zero IT skill, now builds dashboards herself. Together they walked through what belongs on a compliance dashboard, how to balance leading and lagging indicators, why data validation is the foundation of trust, and how to avoid building the Death Star. Here are the key takeaways.

Key Takeaways

Dashboards Beat Static Reports

• Pivot tables are useful, but they require background knowledge to interpret. A dashboard built around core metrics refreshes on its own, so viewers know exactly what they are looking at without a lengthy explanation every time.

• Real-time tracking shortens the lag between when data is created and when someone actually puts eyes on it, such as seeing a substantiation rate shift as soon as cases close.

• Caryn described the old annual training process: copy raw data, filter it by region, paste it, email it, and repeat every week. If she was on PTO, leaders simply didn't get it. Now a continually updated dashboard goes straight to each hospital's inbox, and compliance is no longer the bottleneck.

• The result has been a shift in perception. The compliance department is increasingly seen as a valued resource for the entire organization.

Choosing Metrics That Earn Their Place

• Caryn groups candidate metrics into three buckets: metrics tied to regulatory risk, a mix of leading and lagging indicators, and metrics with a clear call to action.

• Start with the audience. Who is the dashboard for, and what are they supposed to do after seeing it?

• Follow a simple rule: one dashboard, one purpose, five to nine metrics maximum.

• Examples of genuinely actionable metrics include overdue compliance training, aging investigations measured against internal and industry benchmarks, and repeat audit findings that signal a need for retraining.

• William showed an intake-method visual that lets viewers click into anonymity rates by region, facility, or department, which can help target training on sensitive issues like retaliation.

Balance Leading and Lagging Indicators

• Lagging indicators are not bad, but living exclusively on that side of town keeps compliance in reactive mode. A healthy program uses both.

• Credentialing is a clean example. The leading indicator is licenses expiring in the next 30, 60, or 90 days. The lagging indicator is licenses that have already expired, which demands immediate action.

• Nick compared it to a fitness plan: going to the gym every day is the leading indicator, and whether the scale moves is the lagging one.

• William noted that the biggest barrier to leading indicators is access to data. Without quantifiable inputs, teams end up looking only at symptoms after the fact.

Garbage In, Garbage Out: Validation Builds Trust

• Begin with a single home location for your data. An initial manual pull may be necessary, but the goal is a continuous feed that eliminates repetitive manual work.

• Validate before you publish, then validate again before you share. William said this is where most of the up-front time should go.

• Publishing inaccurate data costs you momentum and the trust of stakeholders who rely on what you provide.

• Validation is ongoing. End users will never enter data the same way every time, so periodic checks keep the data clean.

Don't Let Perfect Be the Enemy of Good

• No dashboard is ever 100% accurate. Blank fields, typos, and free-text entries that must be translated into categories will always introduce some noise.

• Over time, you build a relationship with your dashboards and learn the normal ranges, so anomalies stand out quickly.

• Be transparent. A short note acknowledging that the data has been sanitized but may fluctuate due to system idiosyncrasies keeps credibility intact.

• Nick added that directionally correct data is often enough to act on. You don't need to know it's exactly 42.6 degrees outside to know you need a coat.

You Don't Need an IT Background

• William's short answer to whether this requires technical expertise: no. You need curiosity, an understanding of your business, and a tool that lets you analyze the data.

• ScionHealth uses Power BI because it fits their Microsoft environment. Data can live in a Teams folder or an intranet site, and building a visual is largely drag-and-drop.

• Caryn recommended sandbox time: spend half an hour a day building dashboards nobody else ever has to see, then wipe them and start again until it feels comfortable.

• Find a partner to bounce drafts off of, and share early versions with a trusted group before rolling them out broadly.

• Nick encouraged attendees to shed the "I'm not a math person" identity. Everyone already does data analysis every day, from choosing a flight to planning a commute.

Design With Psychology in Mind

• ScionHealth's training compliance dashboard is shared openly across leaders, with completion percentages by facility and drill-downs to department, leader, and individual employee. Visible results naturally motivate leaders to follow up with their teams.

• Use color with intent and keep it consistent. Red means stop or pay attention, green means good to go, and a metric that is purple this week stays purple next week.

• Consistency trains your audience. The first dashboard might take 30 seconds to understand. By the third, recipients know where to look in three seconds.

• Context belongs in the words, too. A title like "Anonymous Reporting Rates, Q3 2026" with a year-over-year comparison and a national benchmark is actionable. A bare "42%" is not.

• Dashboards should lead to the right questions, not jump to conclusions. Rising anonymous reports could signal fear of retaliation, or a healthier culture where concerns are raised directly instead.

Don't Build the Death Star

• Once a dashboard reaches 10, 15, or 20 metrics, it no longer has a single purpose, and the audience tunes out. Visual clutter hides the risk the dashboard is supposed to reveal.

• It is fine to have 20 dashboards on 20 topics. Pull up the investigations dashboard for an investigations meeting and the corrective action plan dashboard when discussing those.

• Keep a dashboard dictionary, a repository that lists every available dashboard with a short description of its purpose. Often 90% of what you need is already built and just needs a tweak.

• Use drill-downs from division to region to facility to leader instead of cramming everything onto the first view.

• Behind the scenes, a single source file can feed 10 to 15 dashboards, so one refresh updates them all.

• Nick framed it plainly: expertise is knowing what matters and what doesn't. The best directors know which scenes to cut and still tell the story.

Tailor the View to the Audience

• Board and executive dashboards should be high level, with minimal detail, a clear risk story, and comparisons to prior periods and national benchmarks.

• Leadership views focus on trends and risks over monthly, quarterly, and annual timeframes, along with peer comparisons across regions.

• Operations views are task-focused, such as unit-level trends and what regulators are currently scrutinizing.

• Different people take different things from the same dashboard. On a quarterly hotline report, an ethics and compliance officer may notice that staff only use the hotline, while a CEO wants to know the top issues and what is being done to prevent them next quarter.

• Avoid conflicting versions of the truth by using one underlying data source and changing the focus, not the data. Facility leaders may see intake, timeliness, and documentation, while operators look at substantiation and corrective action plans.

Help Leaders Help You

• Leaders want to feel competent when holding their teams accountable. A clunky, manual report undermines that before you even start.

• When a dashboard is genuinely useful, leaders gravitate to it because it lets them speak confidently to their facility leaders and staff.

• Stay available for questions, such as explaining that a training completed after the last data pull won't appear until the next refresh.

• Keep the risk front and center and repeat the update on a steady cadence so progress is visible over time.

• The clearest sign of success: leaders start asking for more dashboards.

Closing Thoughts

Asked for one step attendees could take this quarter, William offered two in one: review your data for accuracy on a regular cadence, and confirm that each dashboard still meets the need it was built for. If it doesn't, go back to the construction zone, gather feedback from stakeholders and operators, and keep iterating until the tool earns its buy-in.

The session closed on a simple principle. Dashboards should start conversations, not end them, by answering what is happening, why it matters, what changed, and what action is needed and by whom.