I wonder what the business model is though. As said on other threads, these machines need maintenance and are not vandalism stable. You can't just put it near the toilets and forget about them. Most likely you would put them on the bar with the bar tender pulling cash out every now and then.
Ok, so now there is this machine. I own a bar in NH. Why would I want it? Commission is 1%. Turn over is $1000 per month (as long as I'm the only one in town) and the machine costs $2000 with internet and electricity? That would mean the commissions pay for it in 200 months? Which bar would do that now? Ok, in New York it would have a turn over of maybe $10k paying for itself if 2 years but which bar would believe that? Sure the makers of the BATM dream of selling thousands but I guess they are ahead of time with this. Either they pay bars to be allowed to put the machine there and earn the commissions or the bars are not ready for that yet.
The business model is to manufacture and sell machines to merchants. They can set the commission to whatever they want, and they keep it all. We are not looking to be a service provider. The motivation for us is to promote Bitcoin globally so our own investments in Bitcoin appreciate in value and, secondarily, to turn a profit on selling machines. The motivation for the merchant is to earn a continuous profit stream on the commissions.
$2000 would be on the high end of the price range we would ask for the machine. We're thinking more in the $1000-1500 range. Electricity usage is minimal: it draws around 40 watts (most of which is for the screen backlight). It doesn't need a dedicated Internet connection; most places that would install this would already have Wi-Fi on site. We'll also have a 3G option, and that will of course have a monthly service fee from a cellular service provider.
The initial target audience is retail establishments that already want to accept bitcoins from customers. Customers who inquire and learn about Bitcoin will naturally want to know where they can acquire them. Such merchants would love to be able to tell their Bitcoin-interested customers that they can purchase bitcoins on site from a machine. Such a feature would establish a repeat-customer loyalty for the merchant as well.
$1000/mo. throughput would be ludicrously low. We put through almost $5500 just at Liberty Forum, which was one weekend. Granted, the Bitcoin adoption density at Liberty Forum is higher than anywhere else in the world (except maybe PorcFest), but adoption is increasing all the time. By the time these machines are ready for mass production, there will be areas with high enough adoption density to make a compelling business case for installing a machine in a high-traffic location.
Ok so the liquidity would come from the bar running this machine, not from your service? That is a huge maintenance bonus.
Ok so it is for places that accept bitcoins and want an easy answer to where to obtain bitcoins? Then maybe they set the fee to 10% just to offer some solution but not to attract people from a 200 miles radius to buy cheap coins. This way it makes sense to have it in a shop that accepts bitcoins although it is a bit artificial to buy the coins I'm intending to spend in the same shop where I could pay with $$ for x% less directly. Well, it's new so people will use it.
So you take the $5.5k/week at Liberty Forum as a reference value with all the geeks and reporters that tried it out just to know what it was? And want to sell thousands? In Germany there is exactly one shop I know of that accepts bitcoin and I am almost 100% sure he is not getting $5.5k worth of bitcoins per month. I know in the USA there are some more shops, but thousands? To make a turnover of thousands each? No way. BitPay bragged about having sold their solution to n customers and days later they bragged about having processed 2n purchases. Sorry I don't remember what n was exactly but 2 purchases per webshop is exactly nothing and that's about what I would expect if you managed to bulk sell these BATMs to McDonalds for example. Having one in every major city, yes, that would work but not thousands. Not this year and maybe not next year neither. I hope you proof me wrong.