The Future Of ATM Software?

I recently attended The Payments Knowledge Forum in London.  This annual gathering of users, suppliers and consultants in the transaction switching business has a 30 year history.  It started as EBUG – The European BASE24 Users Group – before becoming a supplier-agnostic event after BASE24 owner ACI ended its involvement. 

The result is a highly focussed event where the industry can compare and contrast the messages from all of the key suppliers of switching solutions, ATM software and supporting solutions for connectivity, monitoring and testing.  The event is unlike a typical trade show because it hosts roundtable events where competing vendors debate key issues raised by the audience.  These sessions are very effective at establishing the vendors’ positions regarding the issues and they allow the users to compare those positions directly. 

This years’ event included such roundtables on a number of topics. Having participated in one myself (on the topic of cloud-based testing) I can confirm that they are challenging for participants.   As an audience member, I found the ATM Smart Client Roundtable particularly interesting.

The format of the ATM Smart Client round table was slightly different in that it involved two different tier one banks presenting their respective projects together with their chosen vendors. This gave the audience a clear side-by-side comparison of two very different approaches to the ATM Smart Client proposition.  As Wincor Nixdorf and Auriga, the two vendors involved, have the largest global installations of Smart Client solutions, this comparison can be considered a valid survey of the market.

I wasn’t alone in my interest – the audience feedback rated the Smart Client session as one of the most interesting of the conference.  It appears that the PKF committee have found a winning formula for showcasing leading technology – I expect other events will adopt this approach in time.  PKF Operatiosn director Denys Whitely said,

‘This blend of user experience and vendor expertise showcased in a highly interactive set of short presentations and discussions works to the advantage of all parties. It is a neutral, informative, lively and engaging way for us to meet our information-sharing objectives.

During the session, a Barclays Bank representative described how Barclays had first embarked on a multi-vendor software solution project in 2009.  In 2014 they had reviewed this strategy as at that time they still managed multiple different software stacks in their ATM fleet.  In order to accelerate their momentum towards the benefits of a multi vendor solution, Barclays reported that they picked Auriga due to its proven vendor independent ATM solution and its willingness to provide onsite development resources for the duration of the project.  Although ATM software is considered “Off The Shelf Product”, the level of integration required at each site often demands development support to make a deployment successful, regardless of the vendor selected.

Barclays described how they identified the legacy NDC+ protocol and its aging implementation in their BASE24 switch as the biggest obstacle to progress.  So they started by deploying Auriga’s WWS ATM solution with a Smart Client loaded on the ATM connecting to a Channel Manager component in the datacentre.  Then the Bank was able to replace their BASE24 NDC device drivers with a standard, maintainable ISO8583 connection to Auriga’s WWS Channel Manager. In Barclays’ view this has been an easy, low risk step that allows Barclays to have a single Smart Client version running on all of their self-service fleet, managed by a single Channel Manager which handles the relationship with the transaction switch for host and card scheme connectivity.  In turn their customers will enjoy a radically enhanced experience with a modernised look and feel and a rich transaction set.  The Bank reports that the project has gone very smoothly and the first machines are scheduled for go-live before year end.

By contrast, HSBC stated that it had decided to first convert all ATM clients to Proflex4, leaving their legacy (NDC) infrastructure intact. In this way, HSBC is taking advantage of the Proflex GUI tools to quickly change the ATM workflows and to create an enriched customer experience without making any infrastructural improvements to their ATM environment in the short to medium term. HSBC will attempt to address the Channel Manager opportunity at a later date, but have no committed timetable in mind. 

It seems to me that the fundamental difference in these two approaches is in their attention to what I would term the Channel Manager.  In Barclays approach, Auriga’s Smart Client architecture, the WWS server, is seen as the primary point of integration with the rest of the enterprise’s systems.  By contrast HSBC is at present focused on the PC in the ATM as the critical point of integration.

Arguably, it is not surprising that an ATM hardware supplier would focus its attention on the role of the ATM PC.  Conversely, it is quite logical that an independent software vendor like Auriga would want to boost the role of the Channel Manager.  Both categories of supplier choose to serve the interests of their clients in a different way  What is key for ATM deployers  is which category’s interests align most closely with those of the market place? 

KAL CEO Aravinda Korala said:

“There is no doubt that innovation in the ATM channel will focus more and more on what the software can do, while the hardware plays a supporting role in providing the physical essentials. We are seeing software cycles quicken in ATMs with new ideas being tried and tested, and time to market becoming the key. The debate about thin-client vs fat-client becomes moot in this context – what matters most is how quickly innovations can be brought to market and how best that software can subsequently be managed”

In my discussions with banks and ATM processors, most are focussed on reducing the cost and complexity of managing the central transaction switch as well as the cost of new hardware purchases. They are also striving to  extend the capabilities of their networks through the implementation of new and more complex transactions managed within the Bank’s application infrastructure.  To me, this is a strong indication that the Smart Client/Channel Manager architectural model is what the market is seeking. 

Where a Bank has XFS-compliant ATMs already deployed, the Smart Client/Channel Manager model appears to offer the capability to deliver enhanced services while reducing the maintenance costs of the transaction switch without requiring costly ATM hardware upgrades.

Auriga CTO Carmine Evangelista commented, 

“A Smart Client Channel Management architecture offers banks the best of both worlds - the ability to reduce their hardware costs and unlimited freedom to fully automate their transactions and client services”

Aside from the complexities of software capability and architectural choice, new ATM sales by the established vendors are seeing increasing competition from new, predominantly Asian, providers.  For details of this phenomenon, see a recent article by Felix Kronabetter on ATM Marketplace called “RBR study: What does the future hold for NCR, Diebold and Wincor?”

Research by Marketing theorist Igor Ansoff showed that all markets eventually consolidate into a single Market Leader, with one or two Attackers, and a number of smaller Guerrilla companies that are able to achieve local dominance, but lack the resources to dislodge the Leader and Attackers.  In this context the merger between Diebold and Wincor is just another step along a well-established road.  Which of these traditional ATM giants ends up as the single Market Leader remains an open question - NCR or the newly formed Diebold-Wincor conglomerate?  Without doubt, whichever wins will continue to experience growing attacks from the newer suppliers from Asia, putting continuing price pressure on ATM hardware. 

Against this background of hardware commoditisation and major mergers, the impending architectural transition from ATM Fat client to Smart Client/Channel Manager promises to open up new opportunities for Independent Software Vendors in the market for ATM software.   ATM user organisations will await with interest the outcome of any Wincor-Diebold merger, and the resulting hardware/software strategy and roadmap. In addition to Wincor Nixdorf’s family of solutions, Diebold offer Agilis and mPower as well as the recently acquired Phoenix Interactive Software., each with its own merits. 

NCR by contrast noted recently that they embarked on a software centric strategy several years ago and point to the acquisitions of Alaric and Digital Insight as evidence of this. However, in my experience, banks want greater freedom of supply.  They want vendor independence not greater dependency on a single vendor.  NCR’s strategy makes a lot of sense from a vendor’s perspective – ATM hardware prices have been in steady decline for years and it is only going to get worse. But how are banks going to see this? Inevitably some banks will be comforted by a “one stop shop” and will embrace the NCR vision end-to-end. However, the unstoppable global trend towards vendor independence and greater choice suggests that the majority of banks will probably invest in freedom of choice and implement a multi-vendor (best of breed) infrastructure for their next generation service platforms.  

However, the really interesting players to watch in any market are the Differentiators.  They are those vendors who seek to split the mainstream market and achieve dominance in their own niche.  Following the acquisition of Phoenix, only two names spring to mind in this regard, KAL and Auriga. Both of these vendors have consistently enabled freedom of choice for Banks in their selection of ATM hardware.