Your Dashboard Is Full. So Why Are We Still Guessing?
On behavioral health KPIs, organizational accountability, and why having more data doesn't necessarily mean anyone knows what the hell is going on.
There's a particular kind of leadership meeting that I suspect anyone who's worked in behavioral health will recognize.
Admissions is concerned because census is down. Marketing insists referrals are up. Finance reports that charges have increased, but collections somehow haven't followed. Clinical leadership is frustrated because providers are overwhelmed, patients are missing appointments, and documentation is falling behind.
Everyone has a report to support their position, and somehow, everyone can be technically correct while the organization continues to struggle.
Then someone suggests that what we really need is a better dashboard.
Which is interesting, because I'm not entirely convinced the problem is a lack of spreadsheets.
I've spent a considerable amount of my career developing KPIs, building leadership dashboards, evaluating revenue cycles, and trying to understand how behavioral health organizations can be surrounded by information while remaining remarkably unclear about what's actually happening.
I've worked with admissions data that doesn't reconcile with clinical census, billing reports that tell a completely different story from the financial statements, and leadership teams where three departments have three different definitions of the same metric.
And somewhere in the organization, there's usually an Excel workbook with forty-two tabs, six hidden formulas, and a filename containing the words FINAL FINAL UPDATED.
Nobody knows exactly how it works, but everyone is afraid to touch it.
What I've come to appreciate is that the problem usually isn't that organizations don't have enough data. It's that they haven't established a shared understanding of what the information means, how the different pieces connect, or what anyone is supposed to do when something changes.
And that's an important distinction.
Because collecting information doesn't automatically create insight, and having a colorful dashboard doesn't necessarily mean anyone is making better decisions.
Sometimes it just means we've gotten really good at organizing our confusion.
Before We Talk About Performance, Can We Agree on What We're Measuring?
I have a complicated relationship with KPIs.
On one hand, I genuinely love them. I think meaningful performance measurement is one of the most valuable tools an organization can use to understand its operations, improve services, and make more informed decisions.
On the other hand, I've seen enough poorly designed dashboards to understand why some employees hear the letters KPI and immediately develop a stress response.
And I don't entirely blame them.
The problem is that organizations sometimes become so focused on measuring things that they forget to establish why those things matter.
A key performance indicator should tell us something meaningful about how an organization is functioning and help us determine whether action is needed.
It's not simply an interesting number someone decided to include in a monthly report.
Consider something as straightforward as admission conversion.
One department may calculate conversion based on the percentage of inquiries that ultimately become admissions. Another may calculate it using completed assessments as the starting point.
Both approaches can provide useful information, but they measure different parts of the process.
If marketing reports a 25% conversion rate and admissions reports 60%, leadership might spend half the meeting trying to determine who's wrong.
The answer could be that nobody is wrong.
They're just measuring different things.
And unless everyone agrees on how the metric is defined, where the information comes from, and what the result is supposed to tell us, the conversation becomes an exercise in defending numbers rather than understanding performance.
The same problem comes up with census, provider productivity, attendance, collections, and almost every other operational measure.
Even something as simple as determining how many active patients an organization has can become surprisingly complicated when departments use different criteria for defining active.
Is someone still part of the census if they haven't attended treatment in two weeks? What if they're on an approved leave? What if discharge documentation hasn't been completed, but they're no longer receiving services?
Those aren't particularly exciting questions, but the answers matter when leadership is making staffing, programming, and financial decisions based on the results.
I also think organizations get a little too caught up in finding the perfect industry benchmark.
They want to know what their admission conversion rate should be, how productive their clinicians should be, or what percentage of scheduled appointments ought to be completed.
And while external benchmarks can provide useful context, behavioral health programs are not interchangeable.
An outpatient psychiatry practice, a residential substance use treatment program, and an intensive outpatient program have very different operating models. Their patient populations, payer arrangements, staffing requirements, schedules, and clinical expectations all influence performance.
Comparing them without considering those differences doesn't necessarily tell us much.
Before becoming overly concerned about whether a number matches some external standard, I'd rather understand the organization's own baseline, how performance is changing, and what might explain those changes.
Because a benchmark can help identify something worth investigating.
It doesn't automatically explain what's wrong or what we should do about it.
Sometimes the Census Problem Starts Long Before Someone Becomes a Patient
One of the areas where I think organizations frequently miss the bigger picture is admissions.
When census declines, the immediate reaction is often to increase marketing efforts, strengthen referral relationships, or generate more inquiries.
And sometimes that's exactly what needs to happen.
But I've also seen organizations spend significant money trying to attract new patients without first understanding what happens to the people who are already reaching out.
Imagine an outpatient program experiencing declining admissions.
Marketing reports that inquiry volume has remained steady, maybe even increased. From their perspective, they're doing their job.
But when we begin examining the actual admissions process, a different story starts to emerge.
Some inquiries aren't appropriate for the services offered. Others involve insurance plans the organization doesn't accept. Some prospective patients wait too long for someone to return their call, while others successfully schedule an assessment but never attend.
And among those who do complete an assessment, some never actually begin treatment.
Suddenly, the problem isn't necessarily that the organization needs more referrals.
It's that people are getting lost somewhere between asking for help and receiving it.
This is why I think admissions performance needs to be examined as a connected process rather than a collection of isolated numbers.
It's useful to know how many inquiries an organization receives, but it's even more useful to understand how many are appropriate for the services available and which referral sources consistently send patients the organization can actually serve.
Response time matters, too, particularly in behavioral health.
Someone reaching out for treatment may be frightened, overwhelmed, ambivalent, or making decisions during a crisis. Families may be contacting multiple providers because they need help quickly.
And when an organization takes two days to return a call, it shouldn't be particularly surprised if the person has already found another option or decided not to pursue treatment.
I think this is something we sometimes overlook when discussing access to care.
From an operational standpoint, a missed inquiry may appear as a lost conversion opportunity.
From the patient's perspective, it may represent another discouraging experience trying to navigate a system that's already difficult to access.
Both things matter.
I'd also want to know how often the admissions team successfully reaches prospective patients, how many scheduled assessments are actually completed, and what happens between the assessment and the beginning of treatment.
If people are consistently failing to attend their first appointment, there's probably something worth understanding.
Are appointments being scheduled too far out? Are instructions confusing? Is transportation a barrier? Are patients unclear about their financial responsibility? Is the program actually a good fit for their needs?
Or has the process simply become so complicated that people disengage before treatment even begins?
The same curiosity should extend to referral sources.
A hospital might send a large number of referrals, but if relatively few patients are clinically appropriate, complete assessments, or enter treatment, that relationship may not be producing the value leadership assumes it is.
Meanwhile, another referral partner might send fewer patients who are consistently appropriate for the program and more likely to engage.
Raw referral volume doesn't tell that story.
Neither does the number of clicks generated by a marketing campaign.
And when organizations begin evaluating what they're spending to acquire patients, I think it's important to look beyond cost per lead and consider what those efforts actually produce.
There should also be some awareness of referral concentration. A program that depends heavily on one hospital, treatment center, or referral relationship may be doing very well until that relationship changes.
I don't think every organization needs an elaborate marketing analytics system to understand these things.
But leadership should be able to trace what happens from the first inquiry through admission and recognize where the process is losing momentum.
Because spending more money on marketing without understanding those gaps is a little like pouring more water into a bucket without checking whether there's a hole in the bottom.
You might increase the volume.
That doesn't mean you're going to retain any more of it.
Getting Someone Into Treatment Is Only the Beginning
Another area where I think behavioral health organizations can become overly focused on a single number is census.
Census is important. It influences staffing, revenue, provider capacity, program viability, and countless other operational decisions.
But knowing how many patients are enrolled doesn't necessarily tell us whether a program is functioning well.
A program can maintain a relatively stable census while struggling with inconsistent attendance, high rates of early discharge, inadequate engagement, or scheduling models that don't work particularly well for the population being served.
And those issues matter for reasons that extend well beyond revenue.
Consider an intensive outpatient program where admissions appear healthy, but a significant number of patients disengage within the first week.
If leadership is focused primarily on new admissions, the problem may not be immediately obvious.
But when we start examining what happens after admission, we might discover that patients are struggling with scheduling, transportation, work obligations, or expectations that weren't adequately explained during intake.
Maybe the treatment schedule doesn't align with the needs of the population. Maybe the transition from assessment to programming is confusing. Perhaps patients aren't developing a meaningful connection to the program early enough to remain engaged.
Those are operational issues worth investigating.
And they can have direct clinical consequences.
I think first-week retention is particularly interesting because it can reveal whether the experience patients were promised during admissions actually matches what they encounter once treatment begins.
Attendance patterns can be equally revealing.
If patients consistently miss certain groups, appointment times, or days of the week, I'd want to understand why before assuming that the problem is simply a lack of motivation.
Of course, patient engagement is influenced by individual circumstances, clinical needs, and personal readiness for treatment.
But when several people are struggling with the same part of a program, it might be worth examining the program itself.
If an evening IOP consistently experiences poor attendance on a particular day, for example, perhaps the schedule isn't as accessible as leadership believes it is.
If patients are repeatedly missing medication-management appointments, maybe the available appointment times or scheduling process need to be reconsidered.
And if people are leaving treatment before completing the recommended episode of care, I think we should be doing more than documenting their discharge as "client choice."
That explanation may be technically accurate, but it's not particularly useful.
Did the patient leave because of cost, transportation, dissatisfaction, competing responsibilities, clinical appropriateness, or difficulty engaging with the treatment model?
Those distinctions matter if the organization actually wants to improve retention.
And while we're examining engagement, we also need to be evaluating whether the treatment itself is accomplishing what it's supposed to accomplish.
I've written before about how important I think meaningful outcome measurement is in behavioral health, and this is exactly where it belongs.
An organization should be able to look beyond admission counts and attendance rates to understand whether patients are progressing toward their treatment goals, whether clinical symptoms or functioning are improving, and whether important safety and follow-up responsibilities are being completed.
Depending on the services offered, that might involve standardized symptom measures, functional assessments, treatment-plan progress, safety planning, engagement in continuing care, or other appropriate indicators.
CMS and other quality organizations have developed formal measures addressing certain aspects of clinical care, including depression screening and appropriate follow-up and specific suicide-safety-planning activities for eligible patients. Those measures aren't universally applicable to every program, but they illustrate the importance of defining meaningful clinical actions and evaluating whether they're happening.
Because a financially successful program isn't necessarily a clinically effective program.
And a clinical team can't consistently provide excellent care if the operational systems surrounding it make that work unnecessarily difficult.
Which is why I don't think clinical quality should be evaluated separately from the rest of the organization's performance.
A Full Schedule Doesn't Necessarily Mean We Have the Capacity We Think We Do
Provider productivity is another subject that can become surprisingly contentious.
And having spent years working with staffing models, schedules, and performance expectations, I understand why.
Organizations need to know whether their available clinical capacity is being used effectively. Empty appointment slots, inconsistent scheduling, and preventable cancellations can create financial strain while limiting access for people who need services.
But I think we have to be careful about how we define that capacity.
A provider's total paid hours aren't necessarily the same as their available direct-care hours.
Consider a clinical director who works forty hours each week but also provides supervision, manages clinical incidents, oversees documentation, participates in utilization review, and supports program development.
Expecting that person to maintain the same billable-hour productivity as a clinician whose primary responsibility is providing direct services isn't particularly realistic.
Yet I've seen organizations use productivity calculations that fail to account for the actual responsibilities of the position.
The math may be technically correct.
The expectation isn't.
This is why provider utilization should reflect the time genuinely available for patient care after accounting for the other legitimate responsibilities of the role.
And I think it's equally important to distinguish between available time that gets scheduled and scheduled services that are actually delivered.
A clinician might have a nearly full schedule, but if patients frequently cancel or fail to attend, the number of completed sessions could be much lower.
That's a different problem from having appointment slots that were never booked in the first place.
The distinction helps leadership determine whether it should be examining referral volume, scheduling practices, patient attendance, or the way providers' time is allocated.
Documentation also belongs in this conversation.
If clinical notes are consistently late, it may indicate that an employee needs additional training or accountability.
But it could also suggest that documentation expectations are unrealistic, the EHR is creating unnecessary work, or clinicians simply don't have adequate time to complete their responsibilities.
And when the same problem appears across multiple otherwise capable employees, I'd be particularly interested in examining the workflow before deciding that everyone suddenly developed an aversion to charting.
Staffing vacancies and turnover complicate the picture even further.
One unfilled position can affect patient access, increase overtime, shift responsibilities to other employees, reduce available supervision, and place additional pressure on the people who remain.
Eventually, those effects can begin influencing documentation, attendance, quality, and revenue.
This is why I think workforce data belongs in the same conversation as clinical and financial performance.
Not because employees should be treated like another productivity variable, but because the organization needs to understand how staffing decisions affect its ability to deliver the services it's promising.
There's a meaningful difference between having an employee problem and having a staffing model that doesn't support the work.
And leadership should be curious enough to recognize the difference.
Then There's the Small Matter of Actually Getting Paid
I have a particular appreciation for revenue-cycle data, partly because I've spent so much time working through it and partly because few things illustrate organizational dysfunction quite as clearly as services being delivered appropriately and then not being paid for.
An organization can have strong admissions, relatively stable census, and hardworking clinicians while still experiencing significant financial problems because the processes connecting treatment to payment aren't functioning well.
And the frustrating part is that those problems often begin long before a claim reaches the billing department.
Take insurance authorizations.
A patient might be approved for a certain number of services within a defined period, but if scheduling, attendance, documentation, and utilization review aren't working together, the organization may fail to use those authorized services appropriately or miss the opportunity to request continued authorization when needed.
In other situations, services may be delivered without the necessary authorization, creating reimbursement problems that could have been prevented through better coordination.
Those are operational issues with financial consequences.
Then there's the process of submitting claims.
I think clean claim rates are useful because they can reveal whether information is flowing correctly from registration and clinical documentation through billing.
But the important question isn't simply what percentage of claims are accepted on the first submission.
It's why the others weren't.
Are eligibility details incorrect? Are provider credentials or payer enrollments incomplete? Is there an authorization problem? Are documentation requirements being missed? Is the billing system generating errors?
Each explanation points toward a different operational problem.
And this is where departments need to stop treating revenue-cycle issues as something that belongs exclusively to billing.
If a claim is denied because of missing documentation, that's not necessarily a billing-team failure.
If a payer rejects a claim because the provider wasn't properly enrolled, that may reflect a credentialing or contracting issue.
And if patient information was entered incorrectly during registration, the consequences may not become visible until weeks after the service was delivered.
The billing department may be the first to discover the problem, but that doesn't mean it created it.
Denial patterns are especially helpful when organizations take the time to examine the reasons behind them.
A rising denial rate may indicate problems with authorizations, coding, eligibility, clinical documentation, or payer requirements. But simply reporting that denials increased doesn't tell leadership which part of the process needs attention.
I'd want to know whether certain payers, services, providers, or denial categories are driving the change.
And then I'd want to know who's responsible for addressing the underlying issue.
Because repeatedly correcting the same preventable claim error isn't an effective revenue-cycle strategy.
It's expensive repetition.
Collections deserve the same level of attention.
There's a significant difference between charges, expected reimbursement, payments received, and the revenue an organization ultimately retains.
Gross charges can look impressive, particularly in healthcare, but they don't necessarily reflect what an organization is entitled to collect under its payer agreements.
That's why understanding net collections matters. It helps leadership evaluate how much of the collectible revenue is actually being received after appropriate contractual adjustments.
And then there's accounts receivable.
I think organizations sometimes become overly focused on the total amount of outstanding AR without adequately examining how old those balances are or why they remain unpaid.
Receivables older than ninety days can indicate unresolved denials, inadequate follow-up, payer delays, documentation problems, incorrect balances, or claims that may no longer be reasonably collectible.
The number itself matters.
But the explanation matters more.
Because having a substantial amount of outstanding AR doesn't necessarily mean the organization has a substantial amount of cash waiting to arrive.
Sometimes it means there's a substantial amount of work that hasn't been resolved.
And if leadership wants to improve collections, it needs to understand which portion of that work is actionable, where the process is breaking down, and what changes would prevent the same problems from recurring.
Once again, the numbers become useful when they help explain what's happening rather than simply documenting that something isn't going particularly well.
The Most Interesting Information Usually Lives Between Departments
This is probably the part of performance measurement I find most valuable.
When organizations evaluate departments separately, it's easy for everyone to focus on their own responsibilities and explain performance from their own perspective.
Marketing is responsible for generating interest. Admissions is responsible for converting inquiries. Clinical is responsible for providing treatment. Operations manages capacity and workflows. Revenue cycle handles claims and collections.
That structure makes sense from an accountability standpoint.
But it can also create a fragmented understanding of how the organization actually functions.
Because the patient experience doesn't follow those departmental boundaries.
Someone reaches out for help, receives information, completes an assessment, enters treatment, attends services, interacts with providers, and eventually transitions out of care.
Behind the scenes, a series of clinical, operational, administrative, and financial processes are supposed to support that experience.
And when those processes aren't connected, problems begin appearing in places that may not immediately reveal their source.
For example, an organization experiencing low collections may initially focus on billing performance.
But what if the actual problem is inconsistent attendance, resulting in fewer completed services? What if those missed appointments are concentrated within a program whose schedule doesn't work for the population it serves?
Or what if documentation delays are affecting claim submission because providers are covering additional responsibilities created by staffing vacancies?
Suddenly, the financial problem isn't exclusively financial.
It's connected to scheduling, staffing, clinical workflows, and patient engagement.
The same thing happens with census.
A decline in admissions may initially appear to be a marketing problem, but perhaps the admissions team is struggling to respond quickly because responsibilities are unclear or the available assessment schedule is too limited.
And declining patient retention might have less to do with the quality of clinical interventions than with an intake process that creates expectations the program doesn't consistently deliver.
This is why I think leadership teams need to become much more interested in examining how their performance indicators influence one another.
A clinical department can meet its documentation target while patients are still struggling to access services.
Marketing can exceed its referral goal while admissions remain flat.
A program can maintain census while experiencing poor engagement and high rates of unplanned discharge.
And finance can report increased charges while the organization's actual cash position continues to deteriorate.
If every department is meeting its individual goals but the organization as a whole isn't functioning well, perhaps the problem isn't that employees need to work harder.
Perhaps we're measuring the wrong things, or failing to understand how the right things connect.
And that's a leadership issue.
Not a spreadsheet issue.
A Leadership Dashboard Should Help Someone Make a Decision
I think we've developed a bit of an obsession with dashboards.
I understand the appeal. It's satisfying to see performance summarized neatly, particularly when the information is current, visually organized, and easy to interpret.
There's something reassuring about being able to open a report and immediately see whether important measures are improving or declining.
But I don't believe the purpose of a leadership dashboard is simply to make information easier to look at.
It's to make decisions easier to make.
If a metric shows that performance is declining, leadership should understand what changed, what may be contributing to it, who is responsible for investigating, and what happens next.
And once a change has been made, someone needs to determine whether it actually improved anything.
Otherwise, the dashboard becomes another place where problems are acknowledged without necessarily being resolved.
I've seen organizations become very good at highlighting underperforming metrics in red.
Which is useful, I suppose, if the goal is to make everyone uncomfortable in a visually coordinated way.
But a red box isn't an action plan.
Neither is changing it to yellow because someone believes things are heading in the right direction.
Meaningful performance management requires enough context to understand what the data is showing and enough accountability to ensure that someone follows through.
I also think organizations need to be thoughtful about how frequently different measures are reviewed.
Some information changes quickly enough to warrant weekly attention. Inquiry response times, pending assessments, patient attendance, open authorizations, documentation backlogs, and immediate staffing constraints can all affect operations in ways that require prompt action.
Other measures are more useful when examined across a longer period.
Monthly reviews might focus on trends in census, conversion, patient engagement, provider utilization, clinical quality, collections, denial patterns, and workforce stability.
And quarterly conversations should create an opportunity to look beyond immediate performance and consider whether the organization's broader operating model is working.
Are particular programs financially and clinically sustainable? Are referral relationships becoming too concentrated? Does the staffing model support current services? Are patient outcomes improving? Is the organization actually ready to expand?
I think the important thing is matching the timing of the review to the decisions leadership needs to make.
If an organization reviews urgent access problems only once a quarter, it may spend months discovering something that could have been addressed much earlier.
On the other hand, examining a long-term strategic metric every morning probably isn't going to produce much insight.
It may, however, produce a surprising amount of unnecessary anxiety.
And while we're discussing dashboards, I don't think every meaningful measure needs to appear on the executive overview.
Different departments need access to different levels of detail. Revenue-cycle teams need to understand specific denial patterns and payer performance. Admissions teams need visibility into inquiries, follow-up, assessments, and conversions. Clinical leaders need information about treatment engagement, quality, outcomes, and provider capacity.
Executive leadership needs enough information to understand how those systems are functioning together and identify where attention is required.
The objective isn't to fit every number the organization produces onto one screen.
It's to make the information meaningful enough that people know what to do with it.
If Employees Are Afraid of the Numbers, You've Got Another Problem
There's another part of KPI development that I think deserves more attention, particularly in behavioral health.
How the organization uses the information matters just as much as what it measures.
I've seen employees become resistant to performance tracking, and while that resistance can sometimes reflect discomfort with accountability, I think it's worth examining the reasons rather than immediately dismissing them.
Some employees have experienced data being used primarily to identify mistakes, justify criticism, or hold people responsible for outcomes they have very little authority to influence.
Others may have worked in organizations where expectations changed without explanation, reports contained obvious errors, or leadership seemed interested in numbers only when those numbers could be used against someone.
Under those circumstances, I can understand why a new KPI initiative might not be greeted with overwhelming enthusiasm.
But I don't think the answer is to avoid measurement.
It's to establish a more meaningful and fair approach to using it.
A declining metric should prompt investigation before judgment.
If a clinician's productivity is consistently below expectations, leadership should understand whether the schedule is appropriately filled, whether patients are attending, and whether the clinician has responsibilities that aren't reflected in the calculation.
If documentation is repeatedly late across an entire department, perhaps training, workflow design, technology, or staffing deserves attention.
And if several capable employees are struggling to meet the same expectation, I would be particularly interested in examining whether the expectation itself is realistic.
Because sometimes the organization doesn't have several separate employee performance problems.
It has one operational problem affecting several employees.
That doesn't mean people shouldn't be held accountable. They absolutely should.
But accountability should involve clear expectations, reliable information, appropriate authority, adequate resources, and the willingness to examine whether organizational decisions have contributed to the result.
I've become increasingly convinced that data should make leadership more curious, not simply more critical.
When employees believe performance information is going to be used fairly and constructively, they're more likely to participate honestly in understanding what's happening.
But when the data is primarily used to assign blame, people naturally become more interested in protecting themselves from it.
And that's when definitions become political, reporting becomes defensive, and problems remain hidden until they're too significant to ignore.
I don't think that's particularly useful for anyone.
Especially the patients.
We Probably Don't Need Another 70 Metrics
If there's one thing I'd encourage behavioral health leaders to consider, it's whether they're actually using the information they already collect before deciding they need more of it.
I don't think every organization needs twenty-one KPIs on its executive dashboard, much less seventy.
And I certainly don't believe that creating more reports automatically makes an organization more accountable.
I'd rather see a smaller collection of meaningful measures that leadership understands, trusts, and consistently uses.
For an organization struggling with access and census, that might begin with understanding inquiry response times, assessment attendance, admission conversion, active census by program, and early patient engagement.
For one experiencing revenue-cycle problems, the priorities may include clean claim performance, denial trends, net collections, authorization utilization, and aging receivables.
And for a program facing staffing instability or questions about clinical quality, it might be more useful to begin with provider capacity, turnover, documentation timeliness, patient engagement, and appropriate treatment outcome measures.
The exact combination should reflect what the organization is trying to understand and improve.
From there, the process becomes less about finding more numbers and more about establishing consistency around the ones that matter.
Are they clearly defined? Is the information reliable? Does someone understand what influences the result? Is there a realistic expectation for performance? And when the number changes, does anyone know what they're supposed to do?
I also think it's worth periodically questioning whether the metrics leadership has been using for years are still relevant.
Organizations evolve. Programs change. Patient populations shift. Payer arrangements become more complicated, staffing models are redesigned, and operational priorities change.
The information necessary to manage the organization should evolve with those realities.
Otherwise, we risk continuing to measure what's familiar rather than what is actually important.
And there's nothing particularly impressive about consistently achieving a target that no longer tells us anything useful.
The Actual Point
I realize I've spent an unusual amount of time thinking about performance metrics, dashboards, and the operational systems behind behavioral healthcare.
I have a psychology background, and I've always been interested in understanding why people behave the way they do, how systems influence decisions, and why organizations continue repeating practices that clearly aren't producing the results they want.
Apparently, at some point, that curiosity evolved into a deep appreciation for operational data.
Not exactly the most exciting hobby, but here we are.
What interests me about KPIs isn't the numbers themselves.
It's what they can reveal about the relationship between an organization's intentions and its actual performance.
They can help us recognize when people are struggling to access care, when programming isn't meeting patients' needs, when employees are being asked to manage unrealistic workloads, or when financial problems are being created by processes that nobody has taken the time to examine.
They can also reveal where an organization is doing something well, where improvements are producing meaningful results, and which decisions deserve further investment.
But that only happens when leadership is willing to look beyond the number and understand what it represents.
Because behavioral health organizations are complicated, and most of the challenges they face aren't confined to one person or department.
Access affects admissions. Admissions affects census. Census influences staffing and capacity. Staffing affects clinical quality, documentation, and continuity of care. Those processes influence reimbursement, revenue, and the organization's ability to sustain services.
And every one of those pieces affects the experience of the people trying to receive care.
I think that's why meaningful performance measurement matters.
Not because every aspect of healthcare needs to be turned into a KPI, or because organizations should become obsessed with hitting arbitrary targets.
But because good decisions require an accurate understanding of what's happening.
And sometimes that means being willing to discover that the explanation leadership has been relying on isn't actually correct.
Maybe marketing isn't the reason census is down.
Maybe employees aren't the reason productivity is struggling.
Maybe billing isn't responsible for the collections problem.
And maybe the organization's most significant challenge isn't the one that's been receiving the most attention.
That's where I think data becomes valuable.
It gives us an opportunity to challenge assumptions, identify patterns, and make decisions based on something more substantial than whoever happens to be the most convincing person in the leadership meeting.
Of course, that also requires leaders who are willing to hear an answer they weren't expecting.
And as I've learned, an organization's enthusiasm for transparency can sometimes become considerably less enthusiastic when the information points toward a decision leadership made.
But that's part of the responsibility.
If we're going to ask employees to be accountable for their performance, we should be equally willing to examine whether the systems, expectations, and resources surrounding them are contributing to the results.
And if we're going to describe ourselves as an industry committed to evidence-based care, I think it's reasonable to expect our organizational decisions to be informed by evidence, too.
Which brings me back to those leadership meetings.
The ones where marketing, admissions, clinical, operations, and finance all arrive with their own reports, their own explanations, and their own frustrations.
Maybe the solution isn't another dashboard.
Maybe it's making sure everyone is looking at the same information, understands how their responsibilities influence the larger picture, and is willing to work together to address what the numbers are actually revealing.
Because having forty-two tabs of data doesn't mean an organization has forty-two tabs worth of insight.
Sometimes it just means we've found forty-two different ways to avoid asking the uncomfortable question.
Do we actually understand what's happening here, or have we just gotten really good at reporting it?
And perhaps more importantly, once we understand it, are we willing to do something about it?
