The POS as a Platform

The Point of Sale (POS) as we know it is a very old and closed system. The incumbent in the (hospitality) POS market, MICROS systems was founded in 1978. However the POS has seen very little innovation since then. There are some fundamental issues with today’s POSs:

1. Cost

According to costhelper,

  • Retail or restaurant single-register “starter kits” range from $1,500-$2,500; more elaborate, multiple-station systems with features such as touch-screens, automatic ordering and sophisticated reporting capabilities can cost $15,000-$20,000 and up.
  • IBM has POS systems starting around $1,999-$2,499, but costs can increase up to $4,000 or more per station.
  • Microsoft offers its Retail Management System starting at $1,290 for a single store with one cashier’s lane.
  • QuickBooks sells its Basic POS software for $800, a Pro system for $1,050 and a multi-store version (up to 10 sites) for $1,400; with hardware included, prices start at $1,750.

Unless bought through a reseller, some of these do not factor in the cost of installations, upgrades, deployment or support.

2. Extensibility

Many POS systems do not have a development platform. This severely limits the ability for 3rd parties to come to the POS. OpenTable, Shopkick and Groupon are some of the apps that come to mind that have made their way to the POS. Some time back, my friends and I did extensive research into this area.

It is not impossible to build an app for MICROS, as this Quora thread suggests. But it is not trivial. Radiant, another POS, charges you $25K to get a developer’s license (Quora thread).

Building an app for an exiting POS is non-trivial, if not entirely impossible.

3. Portability

This is not something you can move around easily. An iPad is an example of something that can be moved around easily.

4. Upgrades are not seamless

Like many other hardware+software systems that was built in the 90s, these POSs were not built for easy software upgrades. An upgrade to a POS involves downtime and can rarely be done in-house without the help of the POS vendor or reseller.

5. They Suck

When you walk in to a retail store today, you’ll likely see a very single purpose clunky system with some seriously shady software (that looks like it was built in the 90s).

POS as a Service and as a Platform
Imagine a world where POSs are purely SaaS based and shop owners can pay as they go. This eradicates #1, Cost and #4, Upgrade issues (from above) associated with existing POS systems. I’m also convinced that any web based system (or any software that has been written in this decade) can easily outdo the look and feel of many POSs have today (that solves #5 Suckiness). I’m not sure a pure software POS is the answer, so I won’t get into that. There are many companies that are doing pure web/SaaS POS systems today (VendHQ, Erply, MyMicros, CashierLive etc.) But there is something they are lacking…

Everyone knows what an iPhone app is today – what if this notion of an ‘app’ came to the POS. Imagine an entirely web based POS that had a lightweight HTML/CSS/JS “widget” based app platform (somewhat like the Facebook platform, but not as heavy weight, for starters at least). The POS can enable a marketplace for app developers to build apps that can be easily “installed” on to a POS. Imagine how easy it would be then for Foursquare or Venmo or Paypal or Perkville or a Bill-Splitter app or an E-Receipt app to come to the POS. This would solve #2, Extensibility.

I know of several 3rd parties who will pay a nice premium to get on to a POS (I’m saving you the details here, but I’ve done plenty of research in this area). And these guys are having to do some serious workarounds because of the pain involved in building something for an existing POS. Here’s an example of an iPad that Perkville gave to our neighborhood coffee shop (the founder of Perkville shares the pain of the POS not being as extensible):


Yes, I’ve “gotten out of the building”, in Steve Blank parlance, and spoken to several shop owners (small, big and large) to vet that this is something they’d be interested in. The most common concern that was raised: needing a permanent high speed internet connection, but almost all of them were in favor of something that would help save them money and would help them plan better for the future. 


I wasn’t sold when I first heard of Square. Square only appealed to small/tiny businesses I thought. And these guys are a nightmare to market and sell to. Startups have failed because of the challenge it takes to market and sell to small businesses. But this tweet by Chris Anderson the other day piqued my interest:

Over the last year, Square has been gaining some serious mindshare. For small biz owners, the cost of acquiring the hardware doesn’t really exist. Square has most of the issues that I have listed above solved including the one that I didn’t address since it’s a non-software issue, #3, Portability. If Square does actually plan to get into bigger market segments, I think there’s a LOT of money to be made (according to TechNavio, the Global POS Software market is forecast to reach $3,377 million in 2013 from $2,328.1 million in 2009; thus, growing at a CAGR of 9.7 percent over the period 2009-2013 – and that’s just the software market).

In my opinion, Square is in pole position to be building a POS that is a platform. For the better of companies that want to get on to a POS, I hope this is one of the areas that Square grows into.


15 thoughts on “The POS as a Platform

  1. Ai,

    Appreciate the analysis, I’m in no way affiliated with the guys, but checkout – they make an extremely clever POS and I’m sure they would appreciate any feedback / critiques of their solution.



  2. Ai, great read, and some useful insights for us in the industry (I work for VendHQ). Your comment

    > Imagine an entirely web based POS that had a lightweight HTML/CSS/JS “widget” based app platform

    We are certainly focusing on integrations with other platforms as a key component to our product. Currently we integrate with Xero web accounting. Soon Shopify eCommerce. When our API is released, it opens up the possibilities of a range of other integrations.

    Would you envisage a widget based platform working in tandem, or replacement, to these sort of integrations? Just curious to for your insights really?

  3. Hey Nick,

    I think I see them working in tandem. The reason for my post was to really get the point across the the point of sale can be an amazing platform for 3rd party devs. There are companies that will pay good money to get on to a POS and get access to “the basket” or payment information for discounts or payments or even something more general purpose like restaurant reservations.

    Good luck!

  4. What’s great about this article are the larger questions about product strategy… The legacy-ness of POS has stuck around in many solutions, and in their product management orgs.

    You’re advocating for a “things got to change around here SUCKA” and I agree… That said, one question that begs answering is “why the long cycle of Legacy-NESS?” I’ve owned food service companies as well as worked for restaurant software companies (work for one now that is SaaS and REALLY helps companies do things better. My coffee shop is a customer/user of the software we sell – how often do you hear that? My staff loves the web-based scheduling tool), and I’ve learned that what is out there is a lot of legacy and staffers, owners, managers, etc. keep buying it. Why? If we like the Ipad and stuff, why the legacy? Why is it still “shipped”? The change is coming, but when/what is the tipping point?

  5. What about building a POS on OpenBravo? Do you think this excerpt from their website solves some of the issues you were speaking about?

    100% Web-based:
    ensures easy roll out to new users & sites, with access from mobile devices–with cloud deployment option to minimize complexity and capital expense

    Open Source:
    enables full customer control over time, providing the freedom of an ERP that you can truly make your own

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>