Auto Insurance customers are changing - which means trouble for carriers

"We're not making money on direct, online business" is something we hear from a number of carriers. Some of them are fairly new entrants but there have been reports that top direct marketers are also struggling in their core channel.

The excuses given for this are usally 'too much new competition' or 'the 'good' direct customers are already taken, all that are left are the bad ones'. We would like to propose another theory to explain why it's become so hard to make money selling Auto Insurance directly:  the customers are changing. And changing in two specific ways: they are becoming much better internet users and their understanding of what insurance is has changed.

Increasing internet sophistication is driving more manipulation.
The smart phone revolution has given most Americans an hour or more a day of experience using the internet to solve problems. This is leading to more and more customers engaging in 'wargaming'. Wargaming is our term for when a customer goes to a carrier's site, gets a valid quote, then requests a second quote after changing a piece of information - say the territory or dropping a driver. By doing this a couple times, consumers can get a good feel for how an insurer calculates their premium, resulting in more and more well informed data manipulation. We see it happening all the time. And it is costly,

Healthcare Reform (aka "Obamacare") is changing the way that customers understand 'insurance' Customers' concept of  'insurance' also appears to be changing. The traditional understanding of casualty insurance whether it be health, auto or home was that the policy only covered events that happened during the policy life. Preexisting conditions (healthcare) and damage (P&C) were  understood to be excluded from coverage. But with advent of Obamacare the health insurance market has done away with the concept of preexisting conditions. Instead, health insurance covers any health need during the life of the policy. Consumers may now be comparing their health insurance policy to their  P&C policy with its preexisting damage exclusions and asking 'why doesn't my auto insurance cover this?'

Another aspect of Obamacare may be driving consumers to think differently about insurance. Obamacare limited the difference in  health premiums between the sick/old and healthy/young to three times the healthy/young premium. Consumers compare that to  auto insurance's much greater rate disparities and conclude that they're 'unfair'.

Many customers' concept of what is 'fair' is changing in ways that hurt carrier bottom lines. The result may be that many consumers have been persuaded that things that in the past that they would have considered dishonest are in fact the way things ought to be. That filing preexisting damage claims and manipulating data isn't so wrong after all. And with their much better insight into how their premium is calculated, they're in a much better position to do so.

One caution: while the theories we've proposed here are supported anecdotally by the data and interactions that we see in our customer relationships, we do not know of any scholarly study that directly addresses these issues. We're hoping to encourage enterprising academics to take up the challenge of understanding these market changes.

Carriers need to upgrade their fraud identification and resolution capabilities But in the meantime, we are convinced that these factors are driving at least some of the direct auto channel profitability problems and that they they will have more and more impact as time goes by. So it's essential that carriers enhance their abilities to detect, intervene and resolve fraud to counteract this growing problem. And that's what we do.

VeracityID stops fraud before it starts. Our solutions detect, deter and defeat the most frequent and costly auto insurance frauds, during quote, billing, endorsement and at claim.   www.veracityid.com

Insurers can't stop fraud and misrepresentation with only data - Part 1

Insurance carriers use third party data to validate the information auto insurance customers are submitting to get a quote because getting the details wrong about rate-able factors almost guarantees that a policy will be a loser. So carriers use purchased data to test what the customer provides.

While useful, this approach has several pitfalls that can lure carriers into a false sense of security:

The Problem of Data Errors. All data sets have errors: mis-keys, lagging data, missing data, and so on. In the real world, a data set with 95% accuracy is almost unheard of. And third party data sometimes combines multiple data sources which makes things even worse. For example a head of household data set may be merged with a college marketing data set to identify young drivers living in a household. When data vendors create this type of 'synthetic data' the errors multiply.  For example imagine a scenario where two nearly perfect (95% accurate) data sets are being combined to create synthetic data. Combining them drives their new 'Synthetic' data accuracy down to about 90% (.95 * .95). Combine three data sets and it falls even further.

Consumer errors. Most consumers shop for insurance online, even if they eventually use an agent to complete the transaction. This means that consumer errors from mis-keys, misunderstandings and carelessness are frequently introduced into the process. Errors that serve to magnify third party data's error problems.

Fraud and rate manipulation are low probability events. There are many ways that consumers can manipulate data to reduce their premiums or get payments they don't deserve but each of these taken in isolation is a small probability event, a few percent at most. With purchased data having error rates of at least 5 and typically 10% and with consumers making mistakes, the overall number of both false positive and false negative errors on any given data diagnostic can easily equal five or ten times the number of true positives. 

So you can see how impractical it is to use purchased data alone to judge whether a specific customer has submitted accurate data. More work must be done to validate the initial data diagnosis. And critically, most of this work must be done immediately, during the quote session.

It is important to note that this problem isn't particular to insurance. Indeed, if you follow medical research you'll constantly hear "studies show" that this or that food item or activity is healthy or deadly. If you live long enough, you'll hear it described both ways. Using large, complex data sets to diagnose small probability events is inherently difficult, so difficult that government regulators won't allow drugs, diagnostics or medical devices to be marketed simply based upon this form of 'epidemiological' analysis. They require double blind controlled experiments.

Which isn't really possible when quoting insurance online. So what can insurance carriers do to reduce data diagnostic error rates sufficiently to detect, deter and defeat fraud without also rejecting a large proportion of good customers? There are three keys to diagnosing and eliminating misrepresentation and errors in the insurance quote and application process:

• Leverage all sources of data, not just purchased ones.
• Triangulate between  different data types.
• When in doubt, ask the customer.

This approach is described in Part 2 here.

VeracityID solutions detect, deter and defeat the most frequent and costly auto insurance frauds, during quote, binding, endorsement and at claim.

 Learn more at VeracityID.com

Insurers can't stop fraud and misrepresentation with only data - Part 2

In part one (it can be accessed here) we explained that there is a fundamental limit to the ability of data alone to identify small probability events - the error rates inherent in the data and the acquisition process end up dwarfing the targeted events. This results in far more false positive and false negative errors than actual true positives.

What can insurance carriers do to reduce their data diagnostic error rates so that purchased data can become useful? We believe that there are three keys to diagnosing and eliminating misrepresentation and errors in the insurance quote and application process and critically, they must be done during the few minutes of the quote session.

Leverage all sources of data/information. 
There are five different types of data derived from very different sources. 

• Synthetic data is composed of data sets merged together from other data usually gathered by others for other reasons. 
• Harvested data is data that the vendor (or carrier) collects themselves with the intention using it to evaluate customers. As a result it should be more accurate than synthetic.
• Customer data is simply what the customer shares in the quote process. 
• Customer behavior is what actions the customer takes during the quote or series of quote sessions. People often say one thing but are telling a different story with their actions.
•  Finally customer interaction is information derived from a direct intervention with the customer. 

Triangulate between different sources. 

The way to reduce the probability of a false result is to use different types of data from different sources to get a better perspective. For example if synthetic data says that there is a teenage driver but the customer's data says there isn't, then a data capable carrier would triangulate by first looking at the customer's quote behavior: did he get a first quote with the young driver in the policy and then remove them on a second quote? Other analytics could be used to test anomalous relationships between different types of customer data. For example, are there an unusual number of vehicles for the number of listed drivers? Or barring that, a carrier could gather customer interaction data by asking questions driven by the identification of a possible missing driver.

One important caveat. There are often multiple providers of the same data but one provider's data cannot validate a second provider's if - as is usually the case - both data products are derived from the same underlying data sets. This is why getting customer data, behavior and interaction is so important: it's new information.

When in doubt ask the customer.
This raises two questions: First, how can you ask the customers in an automated online quote process? It must be automated. Carriers need the capability to trigger specific automated question/data gathering cascades based upon a specific diagnostic failure. That way the questions are integrated seamlessly into the quote process. It can be done quickly and cost effectively, we're doing it for carriers today.

Second, How does asking the customer about their data make the decision better given it's the customer's veracity that is in question? The answer lies in human nature: people who provide deceptive data know that they are being dishonest. It makes them very uncomfortable when a carrier immediately and specifically asks them about the data they were manipulating. As a result the true manipulators will often abandon the quote. If they don't, the carrier can often take steps to limit liability. For example in the case of the young hidden driver, the carrier simply insists that that driver either be placed on the policy or be explicitly excluded.

The other reason to talk to the customer is all of the false positives. It's quite likely that many of them are customer misunderstandings or mistakes. When these people are asked about the data they don't abandon, they welcome the help to get the right data so they can get a valid quote. By intervening with these customers, carriers help guide them back onto the path to coverage, improving conversion rates while reducing fraud.

Up front fraud and rate manipulation can be managed.
Historically the auto insurance industry has been fatalistic about up front fraud - the dollar value per event was too small and the time frame too short to do much. This is no longer true. Data capable carriers are using data and real time interventions to reduce their premium leakage and preexisting damage claims fraud losses by amounts that - depending on the channel - equal ten to thirty percent of net premiums written.

VeracityID solutions detect, deter and defeat the most frequent and costly auto insurance frauds, during quote, binding, endorsement and at claim.

Learn more at VeracityID.com

The Definition of a Good P&C Insurance Customer

What makes a buyer of personal auto insurance a 'Good' customer? If we define 'good' as 'profitable' over the lifetime of the relationship then there are really two behaviors that characterize them:

Loyalty - the willingness to stick with the same carrier and not churn to the cheapest alternative every year.

Honesty - telling the truth about their risk characteristics up front and filing honest claims thereafter.

Of the two, carriers focus on Loyalty far more than honesty. They routinely track loyalty and have programs to recognize and reward faithful customers. Yet the the value of Honesty is potentially far greater. Or more specifically, the cost of dishonesty is much more expensive per case than that of disloyalty.

This prompts some interesting research questions: 

• What is the value of Honesty?
• How much of the value of Loyalty is really Honesty? In other words, do dishonest customers jump in and then jump out while the honest persist?
• How can carriers legally reward honest customers? (More carrot, less stick)
• How good are carriers in finding the dishonest in the first place?
• If Honesty and Loyalty are highly correlated then should carriers find ways to reward them more?

Let me know if you have any thoughts or reactions because at VeracityID we're focused on finding the answers to these and related questions, for Auto Insurance and the P&C industry overall. 

What P&C Insurers can learn from New York City

Thirty years ago New York City was a mess, the streets were trash strewn and potholed and people were fleeing in droves. There were so many panhandlers, squeegee men, gang bangers and prostitutes that it seemed like they owned the place. Then Rudy Guliani was elected Mayor and named William Bratton to be police chief. Bratton was a proponent of Professor James Q. Wilson's 'Broken Windows' theory. Wilson argued that the disorder and incivility that unchecked petty crime caused bred attitudes and behaviors that made the city's serious crime worse.  He argued that if New York wanted to reduce crime it should start by cracking down on the petty lifestyle crimes: littering, prostitution, panhandling, graffiti. New York City took his advice and as a result is now the safest big city in America.

The Automotive Insurance industry has a similar problem. Our work with carriers indicates that petty 'underwriting' fraud - where customers hide or lie about their true risk profile to get a lower rate - is disturbingly common. While there are some fitful efforts to fight it, petty fraud has historically been a low priority in the industry - the cost gets passed on to policyholders. But like with New York, the pervasive nature of petty underwriting fraud sends a signal to less ethical customers that it's OK or at least not risky to cheat their insurance carrier. And what starts as a little 'white' lie to get a lower rate can erode the moral barrier to more serious fraud.

This less than optimal state of affairs was understandable so long as there was no cost-effective way to find and resolve this type of fraud. But that's no longer true because increasingly there are automated techniques and tools that can identify, intervene and resolve many underwriting frauds during the customer's quote session. My company, VeracityID has pioneered many of them. Their existence means that aggressive carriers can improve their bottom line, get a jump on the competition and reduce the cost of insurance to consumers.

It's an exciting time to be in this business. 

How customers get auto insurers to pay for preexisting damage (and how to stop them)

Say you have some damage to your car: nothing big, maybe a few dents up front with some scratches amounting to a few thousand dollars in repairs. You’d like to get the insurance company to pay for it but you’ve only got state minimum liability coverage. You’re out of luck, right? Not if you’re willing to cheat. This is the story of Mr. Smith, (not his real name) a mild-mannered accountant who hated paying for auto insurance. Mr. Smith was a 'Policy Rocker', a type of fraudster that preys on insurance carriers and their customers.

Every year Mr. Smith would sign up for the state minimum required liability insurance on his cars. Occasionally his cars would have a fender bender or other damage. When the total damage on any one car grew to between $3,000 and $5,000, Mr. Smith would sign up for an expensive low deductible comprehensive auto insurance policy. In the first week to month of that policy he would file a claim for the accumulated damage on the car, claiming a fictitious casualty ‘event’ that just happened to not have a police report.

Now the insurance carriers that Mr. Smith did this to weren’t stupid. They knew that his claims were a bit ‘odd’ but without proof it was just Mr. Smith’s word against their suspicion. Proving anything would have taken time that carriers didn’t have because regulators require that claims be paid rapidly. Besides, the claims were always fairly small, too small to waste an investigator's time on. So the carriers (there was more than one over the years) just paid. His car gleaming like new, Mr. Smith then cancelled the comprehensive policy and return to his low dollar liability policy until the need arose again.

We call his shifting back and forth between state minimum and comprehensive insurance ‘policy rocking’. Mr. Smith was a hard-core policy rocker, rocking back and forth for many years (and for all we know is still doing it to another carrier).

But with the right tools carriers can shut down the Policy Rocker’s game. The key is identifying likely ‘rockers’ during the quote and application process using specialized business rules. When the rules fire they initiate a specific intervention process that contacts the consumer or his agent through the carrier's quote system. The suspected ‘rocker' is then asked to take photos of his vehicles using a specially designed web app sent to his smart phone. This app takes the consumer through a secure imaging process while registering the GPS, Date/Time and other data that makes it hard to spoof. We call it idMobile.

But the Mr. Smiths of the world usually don’t get that far because once they realize they've been caught, they go down the street to brand X. Which is just fine by us. We're VeracityID and we build real time fraud identification and intervention tools to stop this kind of dishonest.

To beat rate manipulation, Property and Casualty carriers should watch their customers shop.

Most large personal lines carriers feature on-line quotation systems where consumers can get a quote in as little as three minutes. This has resulted in an explosion in the number of consumer quote requests per policy issued. This in and of itself isn't really a problem but the ability to request and get multiple quotes in a short period of time is is teaching some consumers how easy it is to manipulate their data to get a lower premium. That 'education' and the much higher quote volumes means quite a bit of fraud is getting written into policies. In Auto up to ten percent of net premiums written.

It turns out that you can learn a lot about a consumer simply by watching their online shopping/quoting behavior. 

It's a big problem but the fraudsters have a vulnerability that carriers can target: a digital 'paper trail' of their quote history. The way that unscrupulous consumers figure out what mix of the truth, dishonest and withheld facts yields the lowest premium is by varying key rating factors in a series of quote requests. So if a carrier can retain and track a consumer's quote history and apply the appropriate analytics, they can usually pinpoint where the consumer is 'cheating'.

This is harder than it looks because to effectively police this type of fraud, carriers must 1) track and compare quotes for the same risk over time, 2) identify the manipulation(s), 3) estimate the premium impact and 4) intervene and communicate with the consumer. All during a 3 to 5 minute online quote session. This can only be done with highly automated and synchronized analytical and intervention tools that can identify, intervene and resolve the deception within the short time it takes to generate an online quote. That's where VeracityID comes in. We've built the industry's only end to end fraud identification, intervention and resolution solution specifically designed to identify and eliminate fraud at the beginning, before a policy is bound and a commitment made.

Auto Insurers keep looking for their lost keys underneath the streetlight.

Root is a new auto insurance product that uses an app to track prospective customers' driving for a period of time. The premium is based - in part - on the experience gathered by the app. This is appealing because if I think I'm a better than average driver, I am likely to download the app - it's so easy. It's a great customer segmentation strategy that allows Root to simply sort through all the app data for the good drivers and offer them a better rate. But I'm not sure it will improve the pricing of risks much. Why? Because we look for lost keys under the streetlight.

Let me explain: In insurance, it is rarely the risk we can see that is dangerous. It's the hidden risks - the ones that we don't see - that wreck profitability. For example, lets say that Root gets data from a husband and wife that own four cars - they're super safe drivers and so they save 20% on their premium. But what Root doesn't know and its app can't tell it is that three of the four cars are used sixteen hours a day in a delivery business. Driven by 19 to 22 year-olds who are paid for speedy deliveries. The better insight into the couple's driving risk is swamped by the much larger risk that was totally missed.

Root has developed a great system to get much more precise risk data on the drivers that the customer wants Root to see. It's like they have a very bright streetlight - and so long as the keys (risks) are in the cone of light, Root has an edge. But (as with most lost keys) the worst risks are beyond the light's reach. And having such an outstanding tool that generates such insight on the risk we see could lead us to over-focus on what we already know fairly well rather than to search for the unknown risk.

It is these risks: the  hidden ones that customers don't want you to see - that we focus on. VeracityID's mission is to enable carriers to identify and correct inaccurate and/or fraudulent data up front during the quote process and during the life of the policy. We look where where the light doesn't shine.

Why do auto Insurers experience so much rate manipulation at point of sale?

The auto insurance industry has a large problem with customer rate manipulation at the point of sale. Industry studies and our own work with carriers indicate that this type of fraud constitutes upwards of 10% of net premiums. Yet by and large carriers don't focus much resource on fighting up-front fraud. I think it's in part because historically they have been too pessimistic about achieving results. Specifically in the past they've assumed:

"There's not enough time" - The modern auto insurance quotation and purchase process is designed to be completed in a matter of minutes, therefore there is no time for the carrier to examine the customer's data and investigate questionable details. If they try, the customer will go someplace else.

"There's no ROI" - The dollar value of each instance of point of sale fraud is small, running from $50 to about $1,000 per policy. It costs more in underwriter investigation time than the potential savings.

"There's too much uncertainty"  - Carriers use third party data to test the validity of various key rating factors like territory, drivers, vehicles and so on but this data is only 80-90% accurate. Carriers don't want to reject good business based upon bad data so they're reluctant to 'pull the trigger' even when the data tells them they should.

But these assumptions are rapidly becoming obsolete and are leading carriers to ignore a significant profit and pricing opportunity.

"There's not enough time" is no longer true because new tools are available that automate the identification and measurement of fraud risks within a quote session's short duration. 

"There's no ROI" is becoming obsolete because these same tools are increasingly able tyo automate the process of challenging data claims and collecting documentary proof at POS, eliminating almost all of the burden historically borne by Underwriting.

Finally,  "There's too much uncertainty" is only true because carriers mistakenly focus on the data rather than the customer. The data is simply a 'tip and lead' that indicates that there might be a problem. Only the customer can tell the carrier whether the tip is true. And they usually do because once a manipulator thinks a carrier has found him out, he typically abandons the quote and goes elsewhere. Customers who aren't trying to manipulate - the 'false positives' - won't run away just because the carrier asks them a question. This means carriers don't need perfect data - the combination of "good enough" data and new rapid customer interaction tools provide far more insight.

The bottom line is that if carriers update their assumptions about the feasibility of fighting POS fraud they'll find a large and very accessible profit and pricing opportunity. How do I know? Because we're already doing it for their competitors. We're VeracityID.