They Agreed Mileage Was the Most Important Data. Then They All Said No.

Fred Blumer’s route into insurance ran through connected cars. At Hughes Telematics, a company he co-founded, his team built connected vehicle systems for Mercedes-Benz, Chrysler, Nissan, and VW — the category OnStar defined, with the Mercedes product called mbrace. Hughes went public via SPAC and was acquired by Verizon in 2012, becoming Verizon Connect.

Along the way, his team designed and built State Farm’s usage-based insurance program, Drive Safe & Save. And that work produced the insight the rest of his career rests on: of all the connected-car data an insurer might collect — speed, location, time of day, day of week, cornering, braking, acceleration — roughly 75% of the underwriting value is simply how much the car is on the road.

His way of putting it: cars in garages do not have wrecks.

The corollary bothered him. Consumers were handing over enormous amounts of personal data for a discount, and in his view where your car sleeps at night is none of your insurance company’s business.

In Episode 132 of InsurTechTalk, Fred and I covered why every major insurer turned down his odometer technology, how Mile Auto prices without surveillance, what Porsche wanted from an insurance partner, and how tariffs are about to work their way through auto insurance.

About Fred Blumer

Fred Blumer is the CEO and co-founder of Mile Auto, a pay-per-mile auto insurance MGA that celebrated its eighth birthday shortly before this recording. Mile Auto works with a single carrier partner and multiple reinsurance backers, distributing through independent agents — both digital platforms and brick-and-mortar agencies — and holds a partnership with Porsche Financial Services, marketing to Porsche owners under the Porsche Auto Insurance brand. The company came out of an accelerator at Georgia Tech and is based in Atlanta. Fred previously co-founded Hughes Telematics, which built connected vehicle systems for major automakers and designed State Farm’s Drive Safe & Save program before being taken public and acquired by Verizon.

The Rejection That Created the Company

After leaving Hughes and waiting out his non-compete, Fred partnered with an electrical engineer to build computer vision and machine learning systems that capture and validate odometer readings for any car — no device required. They filed patents, then took it back to his contacts across the large insurers.

The response was consistent and, in its way, admirably honest:

• The insurers agreed mileage was the most important underwriting data
• They piloted the systems, confirmed the technology worked and was cost-effective
• And they were unanimously not interested — because, as they told him, they did not want their low-mileage drivers to understand they had been overcharged for generations

Fred’s summary of that logic: good for you and your customers, not good for me.

An advisor suggested he start his own MGA instead. His answer was that he did not know how to spell MGA, much less start one. He approached people who had done it before to ask them to teach him; they proposed becoming co-founders instead. Mile Auto launched in 2017.

Pricing Without Surveillance

The product distinction matters, because “pay-per-mile” now covers a range of approaches.

• Most competitors plug an OBD device into the vehicle, which collects odometer data plus speed, location, time of day, and day of week — much of it feeding the rating algorithm
• Mile Auto rates on traditional underwriting elements instead: driving history, prior claims, and credit where legally permitted
• Pricing splits into a fixed monthly base plus a few cents per mile driven

Fred’s privacy conviction is grounded in direct experience rather than principle alone. While building telematics systems at Hughes, the company was repeatedly subpoenaed for data — and not only for accident reconstruction. Divorce cases: where was my husband sleeping on a given night. Child custody disputes: does my ex-wife stop at a bar on the way home with the kids. Sometimes the data helped put people in jail.

That is what convinced him insurers should not hold data that can be subpoenaed for purposes entirely unrelated to insurance.

Competitors as Market Validators

Fred is unusually relaxed about competition, and his reasoning is sound.

• US auto insurance is roughly a $300 billion market with 250 million cars on the road — ample room for segmentation
• When Allstate and Nationwide launched their own per-mile products, it helped Mile Auto by validating the category
• Having regulators already familiar with per-mile rating is itself a benefit for a smaller player

This is the underrated advantage of not being first: the market education and the regulatory groundwork are already paid for by someone else.

The Porsche Partnership

The origin is more accidental than strategic. Coca-Cola, noting that large companies tend to kill startups, convened other Atlanta-based Fortune 500 companies to find better ways of working with them — a program called The Bridge Community. It ran only a couple of years, but Mile Auto was invited to pitch, and Porsche approached them afterward.

Porsche’s reasoning was straightforward: most of their vehicles are low mileage, so per-mile pricing fits naturally. Their offer was to license the brand and market to Porsche owners in Mile Auto’s licensed states, provided the product was adapted for Porsche vehicles and drivers.

What Porsche cared about was the ownership experience — they had concluded that one of the worst parts of owning a Porsche was the insurance experience, and wanted a direct hand in fixing it:

• OEM parts
• Agreed value coverage
• Concierge claims service
• Priority access to Porsche-certified repair shops

Porsche Financial Services stands behind the program, audits Mile Auto annually from Germany, and trained the customer care and claims teams. Fred’s tongue-in-cheek complaint: Porsche requires the entire team to go to the Porsche track once a year and drive, because anyone speaking to a Porsche customer needs to genuinely understand the brand.

He also corrected my pronunciation, which he had to learn himself: it is Por-sche, two syllables, named for Ferry Porsche.

Tariffs Working Through Auto Insurance

We recorded on April 8th, with the situation still unfolding, and Fred was careful to flag his uncertainty rather than forecast confidently.

• Tariffs will hit imported parts and imported vehicles, but who bears the cost — manufacturers, wholesalers, or consumers — remains unsettled
• His expectation is that the burden gets shared across the supply chain, with manufacturers absorbing some to protect market share rather than shocking the market with an immediate 20–25% increase
• The effect on insurers arrives through two channels: parts costs, and rising used vehicle values that increase actual cash value on totaled vehicles
• A 20% tariff will not flow through as a 20% cost increase to insurers — perhaps half that
• His pointed observation: several insurers have recently filed to reduce rates, and may come to regret it

This compounds a problem already visible over the past decade. A minor dent to a bumper is no longer a minor repair once the bumper houses sensors and cameras, and the semiconductor shortages during COVID demonstrated how quickly parts scarcity turns into claims severity.

Autonomy Changes the Product Entirely

Mile Auto holds patents in this area too, and Fred’s framing is that autonomy shifts the question from auto insurance to products liability.

• Once a vehicle drives itself, the question becomes what was controlling it at the moment of the accident — the computer or the human
• Telematics can already answer that: whether the driver grabbed the wheel or touched a pedal
• Mile Auto’s patents cover bifurcating the insurance product based on what was actually driving
• This becomes a live commercial question as the industry moves from level two to level five autonomy, and as robotaxi fleets scale beyond the small designated zones they operate in today

We also revisited the ride-share precedent — the three distinct coverage states for a driver using their own vehicle: driving personally, logged in and awaiting a passenger, and carrying a passenger. Same car, different insurance posture. Autonomy adds another dimension to an already awkward problem.

The Closing Lesson: Persistence, and Pace

Asked for a failure, Fred described a genuine David-and-Goliath episode.

Early on, after patents were filed and NDAs signed, Mile Auto shared an element of its technology with a large traditional insurer. That insurer studied it, worked with them for some time, declined — and then launched a product and technology remarkably similar to what had been shown to them.

What saved Mile Auto was focus and persistence: not being derailed, outlasting the competitor, and using the pressure to make the product better.

His practical advice for startups working with large companies is more useful than the usual warning about NDAs:

• Keep the pace moving as quickly as you can, even though you do not control it
• Large companies rotate people through programs constantly, and new arrivals do not know the history — they may come to believe the startup’s idea originated inside their own company
• Champions get replaced. While you have one, get something concluded

He extended grace to the company in question, which I found notable given the outcome.

Key Takeaways

• Roughly 75% of the underwriting value in connected-car data is mileage alone — the rest is largely surveillance with diminishing returns
• Every major insurer confirmed the technology worked and declined it anyway, rather than surface generations of overcharging low-mileage drivers
• Telematics data gets subpoenaed for divorce, custody, and criminal cases — a privacy exposure most consumers never price into their discount
• Competitors entering a category validate the market and educate regulators, which benefits smaller players rather than threatening them
• Porsche’s interest was the ownership experience, not the insurance economics — OEM parts, agreed value, concierge claims, certified shops
• Tariff costs will be absorbed across the supply chain rather than passed through in full, but insurers filing rate decreases now may regret it
• Autonomy converts auto insurance into products liability, and the underlying question becomes what was controlling the vehicle at impact
• When working with a large company, move fast while your champion is still there — they get replaced, and institutional memory of your contribution goes with them