On this page
What these are
Four complete public Allstate–McKinsey CCPR files—1,610 pages from a larger claims-redesign record released through litigation.
Why they matter
They place roof, fire, water, theft, contents, auto, and casualty claims inside savings projections and performance measurements.
Why they became public
Policyholders challenged the claims system in court. The litigation record exposed how the program described its goals, tools, and results.
Start with McKinsey’s documents
Eight records carry the argument. Every headline opens the complete public file at the exact source page.
The Billion-Dollar Claims Savings Plan
“67% opportunity captured” · “Net savings” · $1.111 billion
Homeowners claims were not outside the savings model. They were one of the places the model expected to find money.
The Same Hail-Damaged Roof: $4,050 Versus $1,570
“Estimate written on identical hail damaged roof” · “CCPR estimate = $1,570”
The damage did not change. The process did—and the lower number became the opportunity.
Five adjusters estimated the same hail-damaged roof. The chart reaches $4,050; the CCPR estimate is $1,570. The page calls the difference “economic opportunity.”
Read the source pageThe Roof Test “Succeeded” by Driving Payments Down
“successful to date in driving significantly lower severity and closed cost”
The test was scored by how far claim cost fell. Lower payment was presented as success.
The Albuquerque roof-test update says the process drove significantly lower severity and closed cost—and that the reductions exceeded the fact-finding projections.
Read the source pageManagers Were Told to Make the Savings Process Stick
“Strictly following the process is essential to capturing the economic opportunity”
Adjusting judgment was placed inside a management system built to capture the opportunity on every claim.
The document calls for file reviews, reinspections, and ride-alongs to make the roof process “stick 100%.” Its reinspection chart measures economic opportunity per claim.
Read the source pageClaims Were Prioritized by “Economic Opportunity”
“Process-specific prioritization based on economic opportunity”
The money question entered before the inspection, the scope, and the estimate.
The proposed dispatch system collected claim information, assigned priorities, and routed losses using economic opportunity before the claim reached the adjusting process.
Read the source pageThey Put a Price Tag on Every Roof
“Economic opportunity per CWA roof” · Non-CAT $472 · CAT $549
Before Miss Jones’s roof became an estimate, the system had already assigned roofs an average savings target.
The proposed roof process assigned an average economic opportunity to each roof and linked it to damage identification, mandatory scoping, and repair-versus-replace tools.
Read the source pageFire-Claim “Success” Meant Reducing What Was Paid
“To measure fire process success in reducing fire severity”
A family’s fire payment became a performance number to drive down.
The fire-process scorecard defines average dwelling-fire severity as total dwelling dollars paid divided by the number of fire claims. Its stated purpose was measuring success in reducing that figure.
Read the source pageThe System Recorded Whether You Had Help
“Insured Representation: 0-None, 1-P/A, 2-Attorney”
Representation was not a footnote. It was a field in the claims system.
The property-claim process screen tracked whether the insured had no representative, a public adjuster, or an attorney alongside scope, estimate, payment, settlement, and closure data.
Read the source pageNo key record matches that search. The four complete source files remain available below.
How a claim became a financial result
The documents connect the beginning of a claim to the number the company measured at the end.
- 1Before inspection
Rank the claim
Dispatch could prioritize losses by “economic opportunity.”
Bates H000001645 ↗ - 2During adjustment
Control the decision
Managers used file reviews, reinspections, and ride-alongs to make the process stick.
Bates H000001216 ↗ - 3At settlement
Drive the number down
The roof test reported significantly lower severity and closed cost as success.
Bates H000001199 ↗ - 4After closure
Measure the savings
CCPR projections converted captured opportunity into net savings.
Bates H000000524 ↗
Delay. Deny. Defend.
That phrase came later. The documents call the program Claim Core Process Redesign—CCPR. The source record shows the economics underneath the phrase.
The proposed roof process puts denial, coverage, damage identification, and repair-versus-replace inside the same economic-opportunity design.
Open the roof process ↗The papers describe layers of triage, assignment, review, reinspection, and management oversight between notice of loss and settlement.
Open the management record ↗The public court record later described hard-line treatment of represented claimants as part of a “zero sum economic game.”
Read Jacobsen v. Allstate ↗Professor Jay Feinman popularized “Delay, Deny, Defend” as a description of insurer claim practices. It was not the title McKinsey used for CCPR.
What the United States Senate heard
In 2025, testimony connected the McKinsey model to a broader insurance-industry system.
Doug Quinn testified that McKinsey taught carriers to turn claims departments from cost centers into “profit centers.”
Watch and read the testimony at 2:58:50 ↗
Senator Josh Hawley described the model as “delay, delay, delay” and a “zero sum economic game” in which the insurer wins.
Watch and read the Senate hearing ↗
Then compare State Farm’s own documents.
The State Farm records stand on their own. They show a different carrier measuring roof outcomes, manager approvals, and claim payments—then reporting “about a $1.4B decrease in indemnity” after its Wind/Hail program went nationwide.
Read all 1,610 pages yourself
Each complete file has a clean reader URL and a direct hosted PDF download. No login. No gate.
CCPR Benefit Projections and Claims Redesign Materials
490 pages · begins at Bates H000000520
Homeowner CCPR Tests, Roof Results, and Management Controls
495 pages · begins at Bates H000001010
Property-Claim Design, Dispatch, Roof, Fire, and Contents Records
462 pages · begins at Bates H000001505
CCPR Property-Claim Process Screens and Implementation Records
163 pages · begins at Bates H000002820
How these documents reached the public
The original files, court opinions, and regulatory record answer different questions. They are kept separate here.
Allstate’s own account of CCPR
In Pincheira v. Allstate, the court reproduced an Allstate vice president’s affidavit describing a top-to-bottom review with McKinsey, process redesign, and nationwide implementation beginning in 1995. Read the opinion ↗
The court’s “zero sum economic game” description
Jacobsen v. Allstate describes evidence presented during class-certification litigation, including early unrepresented settlements and hard-line treatment of represented claims. Read the opinion ↗
The multistate regulatory agreement
The regulatory agreement records the McKinsey engagement and CCPR rollout, the scope of regulators’ examination, and the controls Allstate agreed to add. Open the official PDF ↗
The surviving public PDF archive
The public mirror hosts the four files preserved in this corpus. Open the public archive index ↗
What readers ask first
What are the McKinsey insurance documents?
They are public Allstate-McKinsey materials from the Claim Core Process Redesign program, usually called CCPR. This corpus hosts four surviving public files totaling 1,610 pages and gives the strongest records clean reader links.
Was the program only about roof claims?
No. The files address homeowners claims including roof, fire, water, theft, contents, and catastrophe losses, as well as auto physical damage and casualty claims.
Did McKinsey call the program “Delay, Deny, Defend”?
No. The documented program was CCPR. “Delay, Deny, Defend” is the later phrase associated with Professor Jay Feinman’s account of modern insurance claim practices.
Are these the complete McKinsey files?
They are four complete public PDF files totaling 1,610 pages, not the full historical production. The public archive describes a much larger body of material released through litigation.
Can I download the documents?
Yes. Each complete source file has a clean reader URL and a separate download link. The hosted PDFs preserve the public source pages and add Friedman & Associates attribution beneath the source-page boundary.
If the payment does not match the loss, show us the claim. Send the estimate, payment letter, denial, coverage decision, contents inventory, or business-income calculation. We’ll start with what was documented, what changed, and what was left out.
Show us my claim 877-650-3515