UnitedHealthcare – The Short Version
This is an example of the Free Read document that I offer to organizations.
Two pages pulled from the full review. Public information only. Not investment advice. August 26, 2026.
The margins are fixed. The reputation is not. This company spent 2025 nearly breaking and 2026 repairing itself, and almost nobody outside the building knows how much of the repair is real. That gap is the whole job.
What Is Actually Going On
1. Shrinking on purpose. Revenue guided below the prior year for the first time in over a decade while operating earnings rise roughly a third. They are firing unprofitable revenue and repricing what is left. Nobody has told the customers that.
2. The repair is credible. Medical care ratio down to 86.7 percent, two guidance raises in six months, and reserves strengthened rather than released. Analysts read the reserve decision as genuine confidence, and they are right to.
3. The reforms are real and invisible. Prior authorization down 30 percent, two thirds of pediatric requirements gone, ACA profits rebated to about a million members, rural payment timelines halved, transparent pharmacy pricing. Almost no member has heard any of it.
4. N/A
Do These This Week
5. Correct Wikidata. Two hours, one person. It is the table that feeds machine understanding, and it is wrong today.
6. File the database corrections. One public database still lists UnitedHealthcare as formerly known as Bind and gives two different headquarters cities. Four hours.
7. Mark up the homepage FAQ. The answers are already written and already approved. Only the schema is missing, which means the machines cannot see them.
8. Check the crawler rules and publish an llms.txt naming the pages you want quoted. One day.
9. Fill the alt attributes. The homepage hero ships at 3840 pixels wide with nothing in the tag. Five minutes, unmade for years.
10. Baseline the measurement before touching anything. Thirty prompts, five assistants, log the citation URL every time, so improvement is provable later.
The Ninety-Day List
11. Answer the worst question yourself. Publish a permanent page titled to match “is UnitedHealthcare good insurance,” in plain language, with real numbers and named limits.
12. Publish the claims statistics monthly with a stated methodology. A comparison site’s figure is currently the canonical answer because you declined to publish one.
13. Build the comparison pages against Aetna, Humana, Kaiser and Elevance, and concede where you lose. A page that never concedes never gets cited.
14. Give the reforms one permanent home, dated and versioned, with a quarterly scoreboard by category and state including the misses.
15. Tell physicians what changed. Two thirds of pediatric prior auth requirements are gone and pediatricians find out by trying. Cheapest unclaimed win in the file.
16. Design the holdout test for branded search now so it runs in spring. During an eight week enrollment window every channel looks like it works.
Where The Competition Is Beating You
17. Elevance framed AI as denial prevention and cut prior authorizations nearly 70 percent on two thirds of your budget. Same technology, opposite story, and they are winning it.
18. Aetna owns the first ninety days with fully digital onboarding for four million members. That window is where loyalty gets decided and only one of you is treating it as a program.
19. Humana is buying the title. They may pass you on Medicare Advantage membership with roughly 20 percent of members in four star plans against your 78. Prepare that comparison before the headline runs.
20. Kaiser gets recommended by name. You get tolerated because it is what the employer offered. That distinction is the entire brand problem in one sentence.
Stop Doing These
21. Stop answering with averages. Ninety percent of claims paid on submission does not answer a finding about post-acute denials at three times your overall rate.
22. Stop calling it misinformation. Correct what is factually wrong once, precisely, then stop. Every extra rebuttal turns an operations problem into an identity fight.
23. Stop using lawyers on critics. The defamation firm and the move to block a shareholder vote each generated more coverage than the thing they were meant to bury.
24. Stop describing AI as cost management. Every phrase about administrative burden reads to a hostile audience as automating denials, because it can.
The Three Hires (if not already hired)
25. <HIRED>
26. Head of AI search visibility in Q4, with budget authority over schema, entity records and third party corrections. It is an entity management job, not an SEO job.
27. VP of incrementality in Q1 2027, in time for 2028 planning. Until holdout testing exists nobody can say which marketing spend paid.
The Question Nobody Has Answered
Margin repair finishes as a story by late 2027. What drives growth after that? Every analyst has asked, and every answer so far has been a list of capabilities rather than a number. That absence, not the Justice Department, is what the multiple is pricing. Answer it at an Optum Insight investor day with segment level disclosure and a five year target, and you get the cheapest multiple expansion available to any company this size.

Discover more from 1000 Startups
Subscribe to get the latest posts sent to your email.
Claude Penland builds the marketing and business strategy for companies that are good at what they do and hard to find. Thirty years operating, one exit, eight of them as a practicing casualty actuary.
The free two-page read is genuinely free. Email claude@1000startups.com and I'll send back what I can see from the outside. Or see the work samples and how to work with me.