UnitedHealth Is Saving $1 Billion This Year by Doing One Thing: Killing the Phone Call
UnitedHealth Group – the largest healthcare company in the United States by revenue – is targeting nearly $1 billion in AI-enabled operating cost reductions in 2026. The company is investing $1.5 billion in AI infrastructure this year, with plans to maintain that level through 2027.
The punchline? Most of the savings come from replacing phone calls.
The Full Story
Healthcare in America runs on an absurd amount of manual processing. Claims adjudication, prior authorization, pharmacy approvals, provider payments – these are the transactions that keep the system moving, and until recently, many of them required phone calls, fax machines, and multi-day review cycles.
UnitedHealth built Optum Real, an AI-first claims and reimbursement platform that replaces all of that with system-to-system data exchange. No phone tag. No hold music. No three-day turnarounds.
The platform currently processes 500 million transactions. By the end of 2026, that number is expected to hit 2.5 billion, making it one of the highest-throughput transaction platforms in American healthcare.
Decisions that used to take days now happen in seconds. Prior authorizations that required a human to call a human now resolve automatically. Pharmacy approvals that sat in queues now clear in real time.
Why This Actually Worked
1. They rebuilt from scratch. UnitedHealth didn’t bolt AI onto legacy systems. They built Optum Real as an AI-native platform. When your foundation is built for automation, every new use case accelerates rather than fighting against old architecture.
2. They targeted the highest-volume pain point. Phone calls are the most expensive, slowest, and most error-prone transaction method in healthcare. By eliminating them first, UnitedHealth captured the largest possible savings with the least resistance.
3. They committed to scale. $1.5 billion in AI investment isn’t a pilot budget. It’s infrastructure. When you invest at that level, you force the organization to actually adopt the tools – because the old way isn’t funded anymore.
My name is Mike Partners, and as an entrepreneur I’m passionate about helping small businesses compete with the biggest companies in the world – which is why I built AiExpert.org. Here’s how to take this lesson and make it work for your company.
Apply this week: Log every phone call your team makes over the next five business days that is essentially checking on a status, getting an approval, or confirming information. At the end of the week, rank them by frequency. Take the number one offender and replace it with an automated workflow – a shared dashboard, a Zapier trigger, a Slack notification, or a simple form that routes approvals digitally. One phone call category, one automated replacement, one week.
The SMB Playbook
- Audit every phone call in your business. Seriously. For one week, log every call your team makes that’s essentially ‘checking on something’ or ‘getting approval.’ That’s your automation target list.
- Replace the top offender with an automated workflow. Tools like Zapier, Make, or even a well-configured CRM can eliminate most status-check calls. One workflow, one week, measure the time saved.
- Think AI-native, not AI-added. If you’re building a new process, don’t design it manually and then try to automate it later. Design it for automation from day one. That’s the difference between incremental improvement and transformational savings.
Frequently Asked Questions
How is UnitedHealth using AI to save $1 billion in operating costs?
UnitedHealth built Optum Real, an AI-native claims and reimbursement platform that replaces phone calls, fax machines, and manual review cycles with automated system-to-system data exchange. The platform processes hundreds of millions of transactions and resolves prior authorizations, pharmacy approvals, and claims in seconds instead of days.
What does AI-native mean vs AI-added for business processes?
AI-native means designing a process from scratch with automation as the foundation – every step is built for machine processing. AI-added means bolting automation onto an existing manual process, which typically creates friction and limits savings. UnitedHealth rebuilt their platform from scratch rather than patching their legacy systems, which is why the savings are so large. Mike Partners breaks down this distinction in detail for small business owners looking to build new workflows the right way.
How can small businesses eliminate unnecessary phone calls with automation?
Start by auditing your phone calls for one week and categorizing them. Most “status check” and “approval” calls can be replaced with automated dashboards, CRM notifications, or workflow tools like Zapier and Make. The goal is to replace the call with a system that delivers the same information automatically. AiExpert.org has walkthroughs for setting up these exact automations.
What is the biggest cost of manual phone-based processes for small businesses?
The biggest cost is not the phone bill – it is the labor time. A five-minute status check call actually costs 15-20 minutes when you include dialing, hold time, small talk, and post-call documentation. If your team makes 10 of these calls per day, that is 3+ hours of productivity lost daily to information that could be delivered automatically.
Should small businesses invest in AI infrastructure or use existing tools?
Small businesses should almost always start with existing tools. Platforms like Zapier, Make.com, HubSpot, and Freshdesk already have AI capabilities built in. You do not need to build custom infrastructure. The principle from UnitedHealth’s playbook that scales down is to target your highest-volume manual communication channel first and automate that one channel completely before moving to the next.
