Earlier this year, a North American tenant protection program administrator walked us through his loss run. His pure loss ratio to the fronting carrier sat at 10%, a quite favorable number on the surface. But then he told us the part that doesn’t make it into the renewal deck: his claims adjusting expense was running at roughly the same figure. For every dollar the carrier paid a tenant, the program paid close to another dollar to have the claim handled.

We hear a version of this from almost every self-storage program we speak with, and we think it is the most important number in the business. It is often a hidden problem because the spend hides in separate places: a per-claim fee on the TPA invoice, a fee for each denial letter, a certificate someone issued by hand, and a loss run that arrives six weeks after the month closed, far too late to be able to react effectively to. Because separate departments are responsible for their own parts, nobody adds them up, and nobody manages them as holistic costs.

This piece is about where that money goes, why existing program structures make the breakeven problem so hard to see, and what a working claims operation built for the self-storage line looks like.

“I spend as much in claims processing as the carrier does paying out claims.”

— Program administrator, tenant protection program

The baseline numbers that need to be fixed:

10% pure loss ratio on the program
Only about ten cents of every premium dollar goes back out to tenants as paid claims. Tenant protection is a low-severity, high-volume line, so the economics of the program are not decided by losses. They are decided by what it costs to handle those losses and how quickly anyone can see them.

~10% of premium spent adjusting those same claims
For every dollar paid to a tenant, another dollar is spent getting the claim to that payment. On most insurance lines, adjusting expense is a fraction of paid losses. Here it is equal to them. That is the breakeven problem the program administrator above is describing: the cost of running claims doubles the cost of the claims themselves, and it is the number a purpose-built claims operation exists to cut.

45 days after month end before the loss run arrives
The administrator learns how the program performed roughly six weeks after the month closes. A facility with a spike in claims, a repeat claimant, or a reserve set wrong has been running unnoticed months by the time it shows up in a report. Slow data is why the first two numbers persist: nobody can fix what they cannot yet see.

The Problem: Fragmented Legacy Systems

Fragmentation is the enemy of efficient performance. There are two key ways that fragmentation shows up as friction in the self-storage claims business: first, at the policy level (much before the claim is ever made), and second, in what it takes for an adjuster to consolidate sources of truth to get the full context that helps them come to a decision.

Slowdown starts with inaccessible enrollment data

The first point of friction occurs at the enrollment level, way before a claim has been made. To understand why, let’s look at the fundamentals of the program structure.

Most programs sit on a single master inland marine policy. Every enrolled tenant is an additional insured on it, meaning no individual policy is ever issued. Enrollments arrive from operators on a monthly bordereau, so the roster the claims administrator works from is always at least 30 days old. Then, one person creates a spreadsheet using that file, and generates certificates of insurance.

So, the data is always lagging. Now, layer in other confounding factors. Maybe the tenant used a direct-to-consumer channel to buy coverage online, forgot the seller's name, and stores at a facility whose name in the claim system does not match any records on the roster. Say they are in Canada, where each province wants a locally licensed adjuster on the file.

This means that every claim opens with questions the adjuster cannot answer from the file alone. Is this person enrolled? At which unit? At what limit? Until someone confirms all of the above, the file waits, and that single wait is where most of the claim expense originates. It is also why a $2,000 water claim can take three weeks to pay, even though the facts were settled in an afternoon.

After verification, fragmentation continues to slow down adjudication

There is a sound logic to how claims are processed at most legacy administrators we talk to. However, when taken together, the steps an adjuster needs to take to close a claim show us why a $325 handling fee attaches to a case that could never pay more than $2,000. It also shows us why tenants become so unhappy with the service that they call the district manager.  

When we break down an example of a very common process, we find issues at every step.

1. Intake does not consistently enforce clean data capture. Instead, there is some version of a three-screen web form with no photos required at filing and no location lookup, leaving the tenant to type the facility name from memory and hope it matches.

   
   

2. Policy verification happens against outdated information. The adjuster emails the program to confirm enrollment against last month's bordereau. Not only is the data already lagging, but the reply comes when it comes.

   
   

3. Follow ups only go as fast as both the adjuster and tenant can connect. The adjuster can only go as fast as it takes to dial each number separately and for the tenant to pick up, resulting in phone tag and diary notes that read “attempted contact” (sometimes pasted in from the previous entry).

   
   

4. Valuation leaves too much to human error. The tenant's list, the adjuster's estimate, a depreciation table borrowed from the carrier (and whatever else may be available) is applied differently by each desk with little accountability.

   
   

5. Approval gets held up with carriers. Often, the carrier isn’t fully comfortable with the administrator and keeps them on low-dollar approval authority. So, the file gets emailed to the carrier and waits on whoever is on vacation.

   
   

6. Invoicing contains inaccuracies. About a third of files go quiet after first notice and should be billed as record-only, but each claim gets billed as full-rate. Telling them apart means counting file notes by hand, so it’s easier to let it go.

Now let’s say the tenant then files a complaint with the Department of Insurance, further complicating matters. The response could take the program a week to assemble. In most cases, the tenant is the one who never picked up the phone, but  nobody can prove it, because the communication lived in an adjuster's inbox rather than in the file. 

The Fix: Put Tenant Protection On a Production Line

In our last paper, we showed how claims work should be split into judgment and production layers. Put simply, manual and repetitive tasks, as well as those that are simply true/false or mathematically objective should be automated, which both reduces error and increases speed. 

Tenant protection is the most production-heavy line we handle in P&C. Theft and water account for the large majority of files, and limits sit between $2,000 and $5,000. The facts are mostly verifiable from data that already exists somewhere: an enrollment roster, a facility address, a weather record, a police report number, and/or a product listing. Consolidating and verifying this data is work a system should do, with a licensed adjuster deciding what the system cannot.

What is done by the system and reviewed by the adjuster

• Enrollment lookup against a live roster or an API into the operator's software

• Facility search with type-ahead across thousands of locations, so the address is always right at filing

• Photos captured live through a link, with metadata, instead of uploads from a camera roll

• Weather verification against NOAA for the loss date and location

• Item matching to a current product listing, then depreciation from the program's table

• Police report and camera footage requests sent to the municipality and the facility

• Acknowledgement, information request, and denial letters drafted for sign-off

• A fraud score that reruns every time new evidence reaches the file

What must be decided by the adjuster

• Coverage under the lease addendum and the policy form

• Whether a fraud flag is real, and what to do about it

• Every payment and every denial

• Department of Insurance complaints and carrier escalations

• The conversation with a tenant who lost family heirlooms

A person is assigned to every file. The system exists to make sure that person opens a file with all the facts already in it.

What Changes When the Production Line Runs

• 48 hours from filing to settlement on most tenant claims

• 48 hours  to load a program's full facility roster into intake, after data cleansing

• 113 checks in the fraud model on every file, rerun as evidence lands

• ~30%  below legacy per-claim rates, with no setup or integration fee

Speed is the unlock. Manual work and consolidating a source of truth across fragmented data sources eat up the economics of a program, leaving little room for savings. By automating the work and moving claims from intake to resolution faster, we lower the cost of handling each claim, fixing the unit economics. 

The economics are only part of the story. The same speed and automation also transform the day-to-day claims operation and its ability to influence the health of the greater organization.

The file is the audit trail

Every claim gets its own email address. The system ensures that all calls, texts, and portal messages land in the file, not in separate inboxes. When a complaint arrives we send the file, and most of the time it goes nowhere, because the record shows who answered the phone and who did not.

Missed tasks escalate before they become letters

A task that is not completed by its due date goes to the manager, then to our VP of Claims. Statutory deadlines are tracked per state, and timeliness complaints do not happen because the system does not allow the silence that causes them.

Record-only is billed as record-only

With less manual tracking, it becomes easy to bill accurately. Claims that stop after first notice are automatically billed at the record-only rate. The first invoice arrives after the first month of claims. 

The loss run is available instantly

With modern data flow practices, a loss run can be a living, always-up-to-date view. It should be filterable by facility, loss type, and coverage level that updates the moment a reserve moves or a payment goes out. This should enable proactivity that prevents new claims from occuring: an administrator can call the district manager about a burst pipe before the facility manager has noticed the water. 

The same view serves the carrier. Anything above authority appears in a review queue with a summary of the file, the adjuster's notes, and the evidence. The carrier approves, approves with a cap, or asks for more, in the file, cutting out long email threads and wait times from out of office contacts.

This is the point at which claims stops being a cost line and starts being the number the rest of the program is run on. Underwriting prices the next renewal from current experience, finance sees reserves as they are set, and the operator sees which units and which coverage levels drive frequency.

Enrollment data can be verified easily with some basic engineering

Most of the expense in this paper traces back to one missing feed: enrollment data. Storage management software already knows which tenant is enrolled, at which unit, at what limit, from what date. When that feed reaches the claims system, the wait at the start of every claim disappears. Verification is instant, certificates issue themselves, and the tenant files from a link inside the same portal where they pay rent.

So long as the data exists, the system can pull it in, no matter how many sources, and consolidate the information needed to get a claim started well before one is ever placed. (And Strala’s engineering team does this completely for our clients).

Raising the Standard of What to Expect from Self-Storage Claims

Challenges across tenant protection programs are relatively consistent and the fixes are straightforward so long as you have the technological ability and manpower. Every item on this list is something an administrator has asked us for in the past year, and every item is something we deliver today.

1. Enrollment verified at intake, from a current roster, not by email a week later

2. A live loss view by facility, loss type, and coverage level, available on the day a claim arrives

3. Photos captured live with metadata, and a fraud score that updates as evidence lands

4. A complete audit trail per claim that can go to a carrier or a regulator without assembly

5. Record-only claims billed as record-only, without anyone counting notes

6. Carrier review and approval inside the file, with authority that grows as trust does

7. Certificates of insurance issued automatically from the enrollment feed

8. Canadian claims handled by desk adjusters licensed in the province, in French where needed

We see these as table stakes, not as nice-to-haves.

The Path Forward

Tenant protection is a good business. Loss ratios in the low teens are the norm, and the frequency is high enough that a claims operation built for the line pays for itself in the first quarter. The waste is not in the losses, but  in the handling, and in the fact that nobody sees the full handling costs until the year is over (making it hard to fix).

We think a program should know its adjusting expense the way it knows its loss ratio, and should expect the two to move apart. This happens when the enrollment feed reaches the claims system, when the production work is done by the system, and when the adjuster's time goes to the files that need a person.

If you run a tenant protection program or the software that operators use to sell one, send us a loss run. We will come back with where the expense sits, what is recoverable, and what the same book looks like on our operation. There is no cost for the analysis.

Schedule a call here.