Good morning!

1. Feature: How to Review Your Business Insurance Policies (4 min)

2. From the Archive:

   • The Hiring Process: Noncompete Clauses and the FTC

   • The Quiet Limits on LLC Ownership

3. Dear TCoL: How to ask AI to determine if we have adequate insurance coverage?

-TCoL

Missed our last feature article? How Passwords Can Kill Your Business

The policy you think you have and the policy you actually have are often two different documents. For many owners, the missing piece is the application you filled out to get that policy in the first place.

When a big claim hits, three documents matter most: the application, the policy, and your contracts. If the application and policy are out of sync, the rest may not matter as much as you think.

The illusion of “full coverage”

If you have ever said, “My agent said I’m fully covered,” you are not alone. Most owners see a declarations page with large limits and a premium they can live with, then assume they have checked the box.

Insurance does not work on assumptions. It works on policy language.

Your insurer prices and issues coverage based on what you told them about:

• What you do

• Where and how you operate

• How many people you employ

• What you own and how much it is worth

If those answers were not accurate, or stopped being accurate, then the policy you have is built on a story that is no longer true. On a bad day, that can be the difference between a paid claim and a very expensive argument.

Why your application is not just paperwork

Think of your application as the “truth baseline” for your coverage. It tells the insurance company who you are and what risk they are agreeing to take.

Common problems only show up years later:

• A startup checks “consulting” as a catch all, then ends up handling sensitive data, building software, or providing design work

• A small shop lists “retail” and later adds installation, on site work, or light manufacturing

• Headcount, payroll, or revenue grows significantly, but the application information is never updated

When something goes wrong, the carrier will often compare:

• What your application says you do

• What your website, marketing, or claim report shows you actually do

If there is a large gap, you may hear phrases such as “misrepresentation” or “material change in risk.” Those are not words you want introduced when you are trying to get a large claim paid.

Coverage gaps you only see when you read

When owners finally sit down with the full policy, not just the summary, several predictable gaps tend to show up:

• Professional work: If you give advice, configure software, design, or manage projects, the general liability policy may exclude “professional services.” That leaves a gap unless you have separate professional liability coverage for your industry.

• Cyber and data: Many policies either exclude cyber exposures or cover them with small sublimits and narrow exclusions that do not match how you actually use technology and vendors.

• Underinsurance: Property, equipment, and business income limits that were fine five years ago may be too low at today’s prices and payroll.

• Auto and employees: Personal vehicles used for business, and employee related claims such as harassment or wrongful termination, often sit outside basic general liability.

These are policy side issues. The application explains why those gaps are there and how the insurer thought you were using the coverage.

How outdated applications quietly create risk

Consider a simple sequence.

Several years ago, you filled out an application describing a small, local operation with a few employees and modest revenue. Today you have more locations, a larger payroll, new services, and contracts with larger customers. Your annual renewals have mostly been “same as last year” with minor changes in limits or carriers.

On paper, you may still look like the smaller, simpler business you used to be. In reality, your risk is now much larger and more complex.

That creates three problems:

• The insurer priced the risk too low relative to today’s exposure

• Some of your new activities may fall outside the way your operations were originally described

• In an extreme case, a carrier might argue that they would not have written the policy on the same terms if they had known the current reality

Keeping your application current on purpose

You do not have to wait for a claim, or even for renewal, to get control over this.

Here is a practical, owner friendly approach:

1. Pull together three things

   • Your current policies

   • The most recent applications or renewal submissions you can obtain from your broker

   • A simple description of what you actually do today, in plain language

2. Compare the story

   • Does the application describe your services, locations, and headcount accurately?

   • Are there lines of business, side activities, or revenue streams that do not appear at all?

   • Has your reliance on technology, vendors, or online sales changed in a meaningful way?

3. Make a short “change list”

   • Write down specific changes since the last full application, such as new services, new locations, major increases in revenue or payroll, new kinds of contracts, and any significant change in how you deliver your product or service

4. Send a concise note to your broker
   Something along the lines of:
   “Here is how our business has changed since our last application. Please confirm whether our current policies are written with these changes in mind and tell me where you see potential gaps in coverages or misalignment.”

It is up to you to give your insurance advisor the raw material to do their job.

Using AI to help

AI tools can be useful here, not as your underwriter or broker, but as your research assistant.

You can:

• Upload the full policy including the coverage forms, endorsements, and exclusions

• Provide a current summary of what you actually do

• Ask where the exclusions and low sublimits might conflict with how you actually operate

Our Sunday Dear TCoL column below walks you through, step by step, how to frame those prompts so you get a focused map of what appears clearly covered, clearly excluded, and potentially underinsured, without pretending the AI can replace a broker or an underwriter.

Treat the AI output as a briefing memo, not a final verdict. The point is to arrive at the conversation with your broker asking sharper questions, rather than trying to replace them.

One quiet afternoon now, less drama later

Insurance is the contract you bought for a day you hope never arrives. On that day, everyone will care about details: how your business was described, what you represented on the application, and what the policy actually says.

You do not need to read every line today. You can give yourself an advantage by doing three things:

• Make sure the story on your application matches the business you run now

• Read enough of the policy to spot obvious gaps around professional work, cyber, property values, auto, and employees

• Use the guidance in this issue’s Dear TCoL column to run a disciplined AI review, then ask your broker clear, written questions about your real world “nightmare scenarios”

One quiet afternoon of honest housekeeping now is worth more than a year of premiums if, when something goes wrong, the insurer pulls out an old application that no longer looks like you.

Dear TCoL: How to ask AI to determine if we have adequate insurance coverage?

Question: My business insurance policy is renewing and I've heard that you can use AI to evaluate policies and determine whether I have good coverage. Can help me with instructions for AI to obtain a detailed review and analysis of our policy?

Answer: Yes, you can use AI to help evaluate an insurance policy and whether it gives you the coverages that you think you have.

Used correctly, it can save time and identify issues that are easy to miss. Used poorly, it can give you false confidence. The difference comes down to how you frame your requests.

Here’s how to do it:

First, start with the full policy, not just the declarations page, and upload it to AI (for insurance policy reviews, we’ve had the best luck with ChatGPT and Perplexity). AI needs the coverage forms, endorsements, and exclusions to do meaningful work. If all you upload is the summary page, the analysis will be shallow and often wrong. If you are worried about privacy, then redact your business information before uploading your policy.

Before you ask for analysis, give the AI context about your business. This matters more than people expect. Tell it what you actually do, whether you give advice, whether you handle customer data, and which types of vendors you rely on. Insurance policies are written broadly, and coverage often turns on how your operations are described.

When you prompt the AI engine, avoid asking whether the policy is good or bad. That invites vague answers. Instead, ask it to map coverage and exclusions to your company’s risk.

A useful opening prompt looks like this:

“I am the owner of a small business. Attached is our current insurance policy. Please review it and summarize what types of claims are clearly covered, what types of claims are clearly excluded, and which risks appear underinsured or ambiguous based on our operations.”

Then follow up with targeted follow-up questions, like:

“Based on this policy, would a claim alleging financial loss caused by advice or work we provide be covered or excluded? Please cite the relevant exclusions or coverage provisions of the policy.

Or:

“If a third-party vendor suffers a data breach involving our customers, which sections of this policy would respond, and what limits would apply to cover our business?”

What AI is good at is pattern recognition. It can spot professional services exclusions, cyber carve outs, privacy law exclusions, and low sublimits quickly. It can also help you prepare intelligent questions or an email for your broker.

What AI cannot do is replace an underwriter or a broker who knows the market. Treat its output as a briefing memo, not a final answer.

One important caution. Do not upload your policy into an AI tool and then blindly follow its recommendations. Use it to understand what to ask your broker.

If you approach it this way, AI becomes a practical tool.

Have an interesting business question and need a free bit of advice? Send your question to [email protected]. No confidential info, please!