The Opportunity for AI in Healthcare

Healthcare is huge. The U.S. spends more than $5 trillion a year on healthcare, or about 18% of the economy.

At a high level, we think about healthcare in four main areas:

• Providers / Care Delivery — organizations that directly care for patients, such as hospitals, clinics, dental groups, and behavioral health providers.
• Payers — insurance companies and health plans that pay for and manage healthcare.
• Life Sciences — companies that make drugs, medical devices, tests, and other healthcare products.
• Healthcare Services & Technology — companies that provide the software, data, staffing, and services that help healthcare organizations run.

Private equity invests across all four areas.

Two areas stand out to us: Providers / Care Delivery and Healthcare Services & Technology.

• Providers / Care Delivery: PE firms have spent years building larger, multi-location healthcare businesses.
• Healthcare Services & Technology: these companies provide the software, data, staffing, billing, and other tools that support them.

Private equity has spent years buying and combining healthcare practices into larger provider groups across areas like:

Behavioral Health · Dental · Dermatology · Ophthalmology · Gastroenterology · Physical Therapy · Urgent Care · Fertility · Orthopedics · Home Health

The model is simple: buy practices, centralize shared operations, and grow across more locations and markets.

But as these groups grow from a few locations to 25, 50, or 100+, they become harder to run.

Each new practice may bring different software, processes, data, and ways of working. At the same time, the same tasks—patient intake, scheduling, insurance checks, prior authorization, billing, and reporting—are repeated thousands of times across the business.

That mix of scale, repeatable work, and operational complexity creates a strong opportunity for AI.

AI can help standardize these workflows, reduce manual work, and make it easier to operate across many locations.

Even across different specialties, most provider platforms follow a similar patient journey:

Patient Inquiry → Intake → Verify → Schedule → Care → Document → Bill → Collect → Follow Up

Behind each step are many smaller tasks.

• Referrals need to be reviewed and entered into the EHR.
• Patient and insurance information need to be collected and checked.
• Patients need to be matched with the right provider and scheduled.
• Prior authorization may be needed before care can happen.
• After the visit, documentation, billing, claims, denials, and follow-up all need to be handled.

Most healthcare platforms already use software for these tasks. The problem is that the systems do not always work well together.

Information may need to move from the EHR to a payer portal, from a phone call to a scheduling system, or from a clinical note to the billing team. Employees are often the ones connecting those systems by hand.

That creates a lot of manual work across the business—and that is where AI can have a major impact.

Today, employees often spend time reading information, moving it between systems, following up with patients or payers, and deciding what needs attention next.

AI can take on more of that work—handling routine tasks, finding problems, and bringing people in when human judgment is needed.

Across a provider platform, we see five areas with the most potential.

Patient access includes referrals, intake, insurance checks, scheduling, and patient communication.

AI can help move a patient from first contact to a booked appointment with fewer manual steps:

Patient Inquiry → Documents Collected → Information Extracted → Insurance Verified → EHR Updated → Appointment Scheduled → Human Review

Instead of staff processing every patient by hand, they can focus on the cases that need extra attention.

Prior authorization often requires staff to check payer rules, gather documents, find missing information, submit requests, and track the result.

An AI-supported process could look like:

Check Requirements → Gather Documents → Find Missing Information → Prepare Submission → Track Status → Escalate Exceptions

AI can handle more of the routine work while staff focus on cases that need judgment or follow-up.

At scale, this can reduce a large amount of manual work across thousands of authorizations.

Revenue-cycle teams manage large numbers of claims, payments, denials, and unpaid balances.

The challenge is not just knowing that a claim is unpaid. Teams need to know why it is stuck, how much it is worth, and what to do next.

Instead of another dashboard, AI can help surface the most important work:

$284K in claims needing attention

$91K likely recoverable

18 highest-priority claims

14 ready for resubmission

AI can review the information, identify likely problems, find supporting documents, and help teams focus on the claims that matter most.

The employee starts with the decision—not the search.

Clinical documentation affects many parts of the business, from patient care to billing and follow-up.

AI can help organize information before a visit, support documentation after the visit, find missing information, and move completed records into the next workflow.

The goal is not to replace clinical judgment. It is to reduce the administrative work around it.

As a provider platform grows, it becomes harder to see what is happening across every location.

AI can monitor performance and quickly show management where problems are happening.

For example:

42 LOCATIONS · 36 OPERATING NORMALLY · 6 NEED ATTENTION

Austin: Patient wait times up 31%. Provider schedules are running behind.

Tampa: Authorization backlog up 171%. Staffing may be the main issue.

Miami: New-patient conversion down 18%. Forty-two after-hours inquiries were not contacted.

Instead of asking managers to search through dashboards, AI can highlight the locations and issues that need attention most.

For private equity, the opportunity goes beyond improving one company.

A behavioral health group, dental group, dermatology platform, and physical therapy business may offer different types of care, but many of their core workflows are similar:

Intake · Scheduling · Insurance Checks · Prior Authorization · Billing · Patient Communication · Reporting · Multi-Location Operations

That creates an opportunity to:

Build → Prove → Adapt → Deploy Again

Solve a high-value workflow in one portfolio company, measure the results, and use what was learned to move faster in the next one.

The exact technology may change across companies. Different businesses may use different EHRs, processes, and systems. But the approach, integrations, controls, and playbook can often be reused.

The first implementation creates value for one company. The playbook can create value across the portfolio.

We are already seeing this in practice through a project involving the migration of patient records between EHR systems.

At first, this may sound like a simple data-transfer task. In reality, it often requires staff to download records, review documents, extract patient information, recreate the patient in a new system, upload supporting files, and then check that everything was entered correctly.

At hundreds or thousands of patients, that becomes a major operational workload.

The manual process looks roughly like this:

Source EHR → Download Records → Review Documents → Extract Data → Recreate Patient → Upload Records → QA

We are redesigning that workflow to look more like this:

Source EHR → Automated Extraction → Structured Data → Validation → Destination EHR → Human QA

In this model, AI and automation handle more of the repeatable work, while people stay focused on validation and exceptions.

That is the larger pattern we see across healthcare: automate the routine path and bring people in where judgment matters.

Graphic to Include

For this section, we should include a simple visual showing:

• the manual KIPU → Valant workflow
• the automated KIPU → Valant workflow
• the time savings and reduction in manual work

The hard part is not coming up with AI ideas. It is knowing which workflows are worth fixing first.

That is why we start with an AI Audit.

1. Map

Understand how the work actually gets done, including the people, systems, handoffs, decisions, and exceptions.

2. Measure

Look at how often the workflow happens, how much time it takes, what it costs, and how it affects revenue, capacity, or patient experience.

3. Prioritize

Focus on opportunities with the best mix of:

Impact × Feasibility × Risk

The best starting points are usually high-volume workflows with clear business value and manageable risk.

4. Build

Put the solution into the real workflow and connect it to the systems people already use.

5. Measure Again

Did costs go down? Did capacity improve? Did turnaround time get faster? Did revenue or patient access improve?

If the business results do not change, the transformation has not created value.

The real value grows as these workflows begin to connect.

Intake → Scheduling → Eligibility → Authorization → Documentation → Billing → Reporting → Operations

At that point, AI is doing more than automating single tasks. It is helping the business run across the full patient journey.

For a provider platform, the opportunity is to solve a workflow once and scale it across many locations.

For private equity, the opportunity is to prove what works in one company and reuse the playbook across the portfolio.

The long-term opportunity is not just adding AI features to healthcare software.

It is changing how the business operates.

We work with healthcare organizations and PE-backed platforms to find where AI can create real business value.

We start by understanding how the business works, where teams spend the most time, and which workflows create the most cost or friction.

The goal is not to leave with a long list of AI ideas.

It is to answer one simple question:

What should we build first, and what value should it create?

Apply for an AI Audit