Auto Insurance Liquidity Crisis
Sometimes we hear about craziness in an industry, and it’s almost too complex to cover. The details make you scratch your head and want to keep asking questions. And, when you finally think you understand what is going on, you realize that you probably don’t, because you still struggle to explain. Then, once you think you can explain it, you wonder what will happen if people were to change. That’s where we are with the world of auto insurance leads.
Before we begin, we should mention a little about auto insurance. There are two main types of auto insurance buyers and companies – those working on a cpc basis and those working on a cost per lead basis. Of the more than $500 million spent per year for auto insurance traffic, it’s probably close to an even split between per click and per lead. Historically, those buying per click didn’t buy much on the lead side and vice versa. For example, Geico will buy targeted clicks to their landing page, but they will not buy any leads. An independent agent on other hand, will buy a lead, because often times the price for the click comes close to the price per lead. Also, they don’t tend to have sites meant to receive traffic. Each model has seen its share of amazing growth and recently its share of challenges. The lead model in particular.
When compared with other mature verticals, agent based auto insurance lead gen has another interesting dynamic – the level of co-opetition. Almost every major player does business with their competitors in order to increase lead liquidity. A company can have 10,000 lead buyers but still have coverage blind spots; that is, they can still have a lead for which they have fewer than the minimum allowable number of buyers. In auto insurance that number is quite high, which has done two things – drive consolidation of lead players but more germane to this story, create a rather robust exchange. This pseudo-exchange, where partial lead data (sans PII) gets submitted to an industry ping tree / exchange for others to bid on and buy.
In a straightforward world, this cloud marketplace means almost any player can have a chance to make more off their leads as well as have a chance to buy leads for whom they have buyers. Here is where the story starts to get interesting. Working with agents directly is really tough, and it takes at least a thousand before a company sees any real benefit. This has kept the number of companies with an agent base low – almost all able to be counted on one hand. Each of these has a traffic side to their business. Almost all rely on some form of affiliate relationships in addition to their internal media buying. The thing about their affiliate leads is that the data is more or less unique to them. Either it’s generated on their own site or they buy data sent only to them.
There is a third, more recent player in this ecosystem, one that doesn’t have its own buyers or generate leads (including working with affiliates); yet, this third party is responsible for the vast majority of activity that takes place in the auto lead ping exchange. If you are wondering, how is it that someone can have no traffic and no buyers yet create a lot of lead traffic? I think that’s a great question. Like any trader, they are simply looking for inefficiencies. No marketplace is perfect, so it stands to reason that there are probably some leads that aren’t finding buyers even though they exist. That leaves room for someone that knows how to play within the cloud. How much room though? While big, the market for these leads that didn’t already find a home isn’t that big.
If leads don’t find buyers, it’s really not because the system doesn’t work, it’s because the leads just aren’t that sellable. In other words, just because a lead can be sold x number of times doesn’t mean it will be. Only a handful of lead types are those that lots of buyers want. So what are you to do sitting within this cloud, without hard assets of your own, trying to grow your business. The answer could be getting good at traffic or getting agents. In actuality, the answer has been to simply modify the leads to make them look better than they are. Please read that again. What some players are doing is getting more leads actually sold by gaming the filters. Perhaps we should be surprised, but every other lead vertical has or still deals with this. I guess what surprises us is that a large percentage of certain companies leads are simply fraudulent.
How do you uncover fraud in the cloud? You are one of the companies with an agent base who sees agents leaving and return rates increasing. And, you try and look at the leads your own system at first didn’t buy but then did buy. And you look at the returns and see if by chance you saw that exact lead slightly differently. What you find might surprise you, and it is here that the auto lead space finds itself, on the realization that way too many leads are almost certainly completely fraud. Now, for the really, really hard part. Will those involved and impacted clean it up? Can they clean it up?
