The Sublime Non-Science Behind an iPad Point-of-Sale System

10977207293_ccbc3da668_kBy Andy Freivogel, Science Retail

A successful niche retailer and I sat down yesterday in preparation for what promised to be an ugly phone call with a point-of-sale (POS) company, whose software was inexplicably failing.

The retailer asked me, “Who are these guys? Are other guys better? Take sign makers, for example. I had a bad experience with a sign maker and I asked someone at the local Chamber of Commerce whether they knew of anyone else. She told me, ‘Well, there’s this guy, and this other guy, but you have the best one. There’s no point in changing. They all suck.’”

The customer looked up at me, paused, and asked, “Do they all suck? Is that the deal with point-of-sale systems?” I wish I could say no, but at some point, they will all suck. It depends on the conditions.

It’s A Crapshoot

We’d like to be comfortable in the knowledge that the technologies at play in the retail space will be quantifiable, absolute, and finite. Ethernet either works or it doesn’t. Routers are either routing or they are not. Music streams when a device has a functioning internet connection. Credit cards can be swiped, and they either return an authorization or a decline. They never say, “I decided not to tell you whether or not you’ll get this money,” or “That didn’t work, and there’s no telling why, just keep swiping. Oops, here are nine authorizations of $7.56 for one card.” Right? Should faith, experience, and unpredictable store conditions logically have any say in a successful transaction between chips and magnets and bits and electrons?

Experienced technicians familiar with that most central of retail technologies—the point-of-sale (“POS”) system—will tell you that whether or not faith, experience, and environment should have any rightful say—they most absolutely have an impact. There are environments in which a particular type of transaction will succeed 99 out of 100 times, as long as the cashier pulls the card through the magnetic reader with the right orientation and at the right speed, and the internet connection and backup telephone line (let’s say it’s Micros or Aloha; no one else is monkeying around with backup analog phone lines today, right?) are working, and the kitchen printer is still powered on and connected to the network, and the plastic grocery bag has been torn in half and inserted in the mag reader slot in order to reduce static buildup and increase friction and surface area. Remove one of those many legs and the entire table topples.

If there cannot be an infallible system out there, can there be POS solutions which get the job done, which are generally reliable, and which can be individually most suitable for specific applications? For example, look at how traffic is managed at any of the 40 or so Portillo’s Drive-Thru establishments. Nine people with baseball caps, pads of paper, walkie-talkies, and change belts are probably the perfect solution for that. Could fewer people with a Square running on iPad Minis accomplish the same thing? Possibly. It depends on the conditions, and one day, those conditions will change. And that’s when stuff will break.

There is a technology for everything out there, and there are few hospitality or Quick Service Retail (QSR) situations in which a problem cannot be solved with technology, particularly if money is no object. Sometimes, however, it is a better investment to train people and enforce procedures than to configure computers and other devices to do these jobs for them.

For example, one customer uses a POS system that is capable of preventing employees from clocking in more than ten minutes early, and will automatically clock them out eight hours later. The customer stresses that this won’t be enough, however, because some employees aren’t scheduled for eight-hour shifts, but rather they are scheduled for four- or six-hour shifts. Now, it’s not impossible to teach the software the conditions under which the automatic eight-hour clock-out doesn’t go into effect. First business rules have to be created, one of which might be “If Employee “A” clocks in at the scheduled start of the shift, refer to Employee “A” Tuesday shift hours to determine auto-clockout.” Once those rules are agreed upon, they need to be coded (maybe in Bulgaria, or India, or San Francisco) and then that code needs to be compiled into a new version of the software running on a server that is providing point-of-sale software to dozens of businesses.

This might work for most of the businesses, but might create a bug or configuration problem for one of them. A month of beta testing would confirm this, and there could still be problems after, as conditions change. Wouldn’t it have been easier to just tell employees to always clock in at their scheduled start time, and then clock out when it’s time to go?


Today’s POS landscape is remarkably different than it was five years ago, and one can safely argue that it is the iPad that has changed the POS market more than any other innovation in that period of time. Define “the cloud” in whatever manner one prefers, but as long as there has been the ability to host data securely in a manner that is accessible to anyone with an internet connection, there has been some version of the cloud and hosted applications.

The size, cost, and portability of iPads are what made them one of the biggest game-changers in the POS realm. The sheer ugliness of the proprietary hardware associated with dinosaur POS systems represented design compromises that made some coffee bars less than what they could have been. Any iPad-driven (and note, some POS solutions are cross-platform and work on Android as well) solution is going to require less counter space. In fact, depending on a store’s philosophy regarding receipts (any good system allows for a sign-on screen along with a choice of tipping prompts), it’s possible that an iPad POS terminal could free up as much as two or three square feet on a countertop. That’s more space for customer engagement, whether it’s artwork, another small digital display to either facilitate the transaction or tell customers what other products/amenities/enticements are available to them, or just space to present a beautifully prepared coffee and pastry. Space to breathe. Downplay the luxury and serenity of such a thing at the risk of your own health and sanity. We’ll run out of space one day.

In addition, the cost of an iPad is always going to be lower than the cost of a proprietary POS terminal. First of all, a cursory internet search for the latest and greatest in the traditional POS ecosystem reveals that new, full-featured touchscreen terminals come in somewhere around $2600. Pour a beer on one, and you’re out that $2600, plus the time it takes to get a technician there (which means lost sales), the cost of the labor to reload software and configure the device (easily $150 to $300) and it will still be obsolete in three years. It can be difficult to gain access to technicians as well. There will be likely be a minimum two-hour wait, one hour to repair (in the best of conditions), and it will cost at least three hours of sales and employee morale. Crack an iPad over your knee, and you’re still only out $500. If it’s running a cloud-based POS software, send the dishwasher to Best Buy with your credit card, load the app when he or she returns, and as soon as you log in, you are back up and running. What did that take, an hour?

A wildly successful coffee bar can serve around 60 people in an hour. The average transaction is going to change from market to market, but in Los Angeles it might be something like $7. Upwards of 65 percent of those transactions are going to be credit card transactions. So, of the $420 in sales that a coffee bar might do during one hour, a broken POS means that 65 percent of those transactions might not happen if there is no way to use credit cards. Half of those customers will have some kind of cash, so let’s just assume that 30 percent of $420 is walking out the door once they find out the POS is down: that’s $126 per hour lost when the POS isn’t working, and this formula doesn’t consider the impact on customer loyalty, word of mouth, and “badvocacy” on Twitter and other social media channels. The next time you visit a coffee bar where they tell you the POS isn’t working, hop on Yelp the next day and count the one-star reviews. This is the best-case scenario when something catastrophic happens to an iPad-based point-of-sale.

It’s important not to confuse a tacit endorsement of iPad POS systems with just any mobile POS or tablet POS systems. There are a few intriguing non-iPad tablet POS systems out there, but in my experience, the ease of replacing an iPad and utilizing the Cloud seems to be the most effective when faced with a POS malfunction. At the end of the day, we all have our preferences, and we’ll support our theories with a combination of experiential anecdotes and cherry-picked internet infobytes.

andy_bioAndy Freivogel is founder of the retail technology consulting firm Science (, which he started after twenty years of work in retail coffee and technology consulting. In addition to guiding specialty and boutique retail customers through the landscape of POS systems, guest wifi solutions, and PCI compliance, he also enjoys free jazz, black metal, skateboarding, street art, and cooking West African dishes. Andy and his wife live in Oak Park, Illinois with their two children, a chihuahua, a pitbull, two cats, and a frog.