3 ms·
As others have pointed out, dealing with fraud at scale is really really difficult. As a result its cost are often priced into sales (either transaction costs o
by ChuckMcM 10y ago
As others have pointed out, dealing with fraud at scale is really really difficult. As a result its cost are often priced into sales (either transaction costs or product costs).
While the privacy implications are large, it is an area where the more data around the transaction can identify systemic fraud. (which is to say organized group that spread their activity across a wide area to keep it below the radar of fraud detection algorithms).
In the case of phones, the seller has a lot of power. They can record the serial number/IMEI data from the box prior to selling it. At which point you can track where the phone is and if its being used. The challenge has always been taking action against the scammers. So perhaps there is a startup idea for a private investigation group in large cities that would go get scammers and recover merchandise. Such action would quickly reveal if the scammers were part of a larger organization or acting independently.
- thebokehwokeh2 10y ago> So perhaps there is a startup idea for a private investigation group in large cities that would go get scammers and recover merchandise. Such action would quickly reveal if the scammers were part of a larger organization or acting independently. You mean the police?
- treebeard901 10y agoNice home you have here, Mr. Amazon Seller. Be a shame if somebody disrupted it.
- ChuckMcM 10y agoI would, if the police were filling that particular niche. As most police departments have limited funds, simple economics steers them to putting their resources on the highest impact cases. So "Some guy defrauded me when {buying|selling} an iPhone." is below the activation threshold for the police. But investigating these "small time" frauds benefits from scale. And like the company that was fixing parking tickets (normally a time consuming and costly exercise for the ticket receiver) by apply resources across a lot of people with parking tickets, I could imagine an equivalent "fraud fighter" business where the "fee" was 1/2 the recovered value. So the company investigates 1000 fraudulent amazon sales in the greater New York area, lets say they can prove half the time that it was fraud and get the appropriate fraud protection coverage to kick in. If they keep 50% of that value, they have now a revenue stream that is 25% of the defrauded value. The mechanic here is that small time fraud allows millions (maybe billions) to be siphoned out of the commercial shopping stream. That burden is generally born by the sellers who raise prices to cover their losses. If a startup can mitigate those losses by 50%, and the seller is willing to split the mitigation 50/50 with the startup, then the seller has 25% less "loss due to fraud" and the startup has 25% of revenue from defeating fraud. It takes money out of the fraud pipeline. When larger fraud rings are uncovered by the startup they can hand them off to the police who are willing to go after multi-million dollar criminal enterprises.