Every wonder why there’s still so much credit card fraud? About 2-3 weeks ago I saw a really documentary on Dateline about online credit fraud, well mostly about identity theft, but still very relevant. It’s what triggered this article. In any case the commentator did a great presentation, and showed how many people get taken. He was even able to show how within a day or so they set up a fake online store to accept stolen credit cards orders on the internet. They did it in a day!
So why is this still happening? Why are stolen credit cards continually used to purchase products when we know how it’s done. Why aren’t credit card companies stopping it cold, or at least bringing it way down?
The answer lies with their motivation and incentives. A while ago I wrote a glowing recommendation on a book entitled “Freakonomics“, where one of the things the book really showed you is that you need to pay close attention to motivations and incentives. And once I started to ponder about the credit card companies motivations, things became a lot clearer. The incentives and motivations just aren’t there to completely remove credit card fraud. There are of course strong motivations to keep it low, but not to completely eliminate it.
Before going into detail, let’s take a look at an example of what happens to us at LandlordMax when someone uses a stolen credit card to purchase our property management software (although it very rarely happens, no one is completely immune to it). Firstly, the fraudster goes online and makes a purchase of LandlordMax. The credit card company authorizes the transaction saying the credit card is valid and has the sufficient funds. Once we get this ok, we then proceed to finalize the transaction and send the customer their product. Everything is looking good and all parties are happy. However within the next few days, more often weeks to months, we receive a notice from the credit company that the transaction they authorized was fraudulent (more often than not it’s because of a stolen credit card). However here’s where it gets interesting, it doesn’t end there, the credit card company then takes back the funds they authorized (remember they told us it was ok to proceed).
So let’s look at the motivations. Firstly they aren’t liable for lost funds. Yes they authorized it, but they aren’t liable, the merchant is. If the card is stolen, the credit card company doesn’t lose a penny, the merchant takes all the risks and losses. Not only does the merchants lose their funds, but you also have to remember that their out of product as well. So for example, if you sold a diamond ring for $2000, not only would you have to give back the $2000 but odds are the diamond ring is nowhere to be found.
Understanding this, it quickly becomes very clear why they aren’t motivated towards a zero tolerance on stolen credit cards. It basically becomes a cost to benefit equation, they have to cleverly balance how much theft they can let go by without people losing confidence in them. To put it in other words, it costs money to detect fraudulent transactions and to be on top of criminals, exponentially more expensive as you get closer to a zero tolerance. What they have to figure out is how much theft is acceptable so that people will still have confidence in their product (and continue to use credit cards) while not paying too much to prevent this fraud. It’s a very fine line to balance. As they get ever closer to zero tolerance their costs go up with diminishing returns.
Remember the key ingredient here is that they aren’t liable for losses they authorized, the merchant is. So their only real motivation is to instill confidence in their customers (credit card holders) that using their credit cards is safe (where they get a percentage of every transaction). They don’t really have a motivation to produce a 100% safe credit card system, at least not until they’re on the hook for the losses instead of the merchants.
Without pushing the point further, don’t get me started about chargebacks. Another one of my favorite pet peeves with credit card companies. If someone ever decides to do a chargeback, of which we’ve had a grand total of 2-3 over the last 4 years which is an incredibly low percentage, not only do the credit card companies take the money back but they also charge you a significant chargeback fee! If you look at their motivation, it’s definitely towards the credit card holders and not the merchants. They have no vested interest in helping you, they don’t lose a cent, they actually make money if chargebacks go through.
So if you take a good look at what motivates credit card companies in terms of protecting people against credit card fraud, its probably not what you expected. Yes they are motivated to keep credit card fraud down, but they are definitely not motivated to completely stop it. It just doesn’t make economic sense for them. Ethically it’s a different story, but unfortunately today the economics is the reality.
Reading many ISV (Internet Software Vendor), uISV (Micro ISV) blogs, etc. you often hear how bandwidth costs can sometimes quickly escalate with success. Jeff Atwood of CodingHorror.com posted an article (with a nice follow-up) on this very subject describing many things you can do to minimze your bandwidth costs. These are:
- Switch to an external image provider.
- Turn on HTTP compression.
- Outsource Your RSS feeds.
All great options, no doubt about it. But another option that is very often overlooked by software companies is the size of their software installers. It’s very easy to forget, never mind completely miss, the size of the installer. For us here at LandlordMax it was always an issue because we also install a local JVM which is considerably large. With each new version the size of the installer steadily increased until it was about 36Mb. This might not seem so big compared to some of the other larger software applications in the market, but when you’re looking at thousands of downloads a month this quickly adds up to a lot of bandwidth.
Late last year we changed the software we used to create our installer to Install4j (you can read my review on this great installer here) and it was able to drastically reduce our installer size by about 50%! Yes, that’s a full 50%! What does that mean? Well if you consider that the majority of our bandwidth is used for downloading LandlordMax, then we were able to reduce our bandwidth by about 50%.
None of the suggestions above could have reduced our bandwidth nearly as much as this one change to the installer. On this blog (FollowSteph.com), the above four suggestions do indeed have a very significant impact, but on LandlordMax because of the nature of the website, the change in the installer completely overwhelmed any of the impact these changes would have had. This is not to say that I don’t strongly believe you should do them, but that in this particular instance another tactic to reduce bandwidth was much more crucial.
So the lesson of the day, other than to do the above great suggestions by Jeff (especially for blog type websites), is to look at your software installer and see if you can’t reduce its size (assuming you’re a software company). The impact may be more significant than you might realize.
In software no good deed goes unpunished
In software there is often a large disconnect between perception and reality. This is a clear example of such a disconnect.
My Own Successful Startup Story … In 1984
A well writen real life story with lots of life lessons.
Top ten things ten years of professional software development has taught me
The title says it all.
Oops! Techie wipes out $38 billion fund
Talk about a disaster. But the real tidbit to get from this piece is if you’re own software/data is correctly (and regularly) backed up?
If Crack Dealers Took Lessons From Walgreens, They Really Would Be Rich
A great example about pricing and complacency.
9 Confessions From A Former Enterprise Rental Salesman
Everyone knows there’s certain things that make renting a car more pleasant. This is the first time I’ve seen it written down.
Features vs Ideology
Ian writes a great article on the debate many software vendors have when including new features into their software.
Good Programmers Do Like to Code; It’s All the Other Crap They Don’t Like
The argument of this article is that developers want to spend time solving problems and not rewriting solutions to the same problems over and over (they don’t want to re-inventing the wheel over and over).
The Truth About the Single Founder Startup
A very good article about what it takes to start your own company if you want to do it solo. This is the route I took and I can attest that you need to wear all the different hats at one time or another.
Why There Isn’t More Multilingual Support
At LandlordMax continually get asked why we don’t support other spoken languages within the software, especially by developers. The truth is that providing multi-language support is no longer that difficult, whats difficult is all the other stuff that goes along with it that Ian describes here.
I just finished all the revenue/sales tallies for last month (March 2007) and it looks like we broke our sales record again! Last month was our highest sales revenue ever for a single month, beating our previous record set in October 2006. It wasn’t a large difference. Actually the number of units sold was exactly the same which in itself is quite remarkable! We broke our sales record within a couple hundred dollars simply because we just shipped a higher percentage of CD’s last month than usual. Like I said, if you look at a difference of a couple hundred dollars over sales in the thousands of dollars, we barely broke our sales record. But nonetheless we did break it (in February we just missed breaking our record by a very small amount which was frustrating).
As for our net total (profits), March definitely wasn’t our record month by any means. We had much higher than normal expenses this month for several reasons which I won’t go into today.
All in all though I’m very excited about March, it turned out to be a great month for us in the bigger scheme of things. And looking at April, if things continue as they are now, we’re looking at another very good month. We’ve already at least matched the number of sales for this time in March today in April!
Well it’s official, we just released version 3.11b of LandlordMax. As I mentioned in my last post, this latest update was probably our biggest update (not including any major upgrades) in terms of effort yet. There were a lot of Vista specific issues that we had to work through for LandlordMax to fully support Windows Vista, some of which even included bugs within the programming language itself! This doesn’t even mention the new hardware and software we had to purchase and setup just to get our test and development environments going.
In any case, we just released version 3.11b of LandlordMax today, so you can now download it. And on top of fully supporting Windows Vista, we also added several smaller fixes for everyone else. You can find the complete list of the latest fixes and changes in the release notes.
Now our main efforts are going to return towards getting the Mac version of LandlordMax ready as soon as possible. Hopefuly that won’t be much longer now too.