James Piggot - Finastra - London 20 August, 2015, 10:59 0 likes

Vishwanath, there are unfortunately banks with ageing mainframe systems that fail such as RBS here in the UK. For them it is very much when rather than why.

Ketharaman Swaminathan - GTM360 Marketing Solutions - Pune 20 August, 2015, 16:45 0 likes

@VishwanathT: Good question:) One that I've asked many times, including in 6 Reasons Why Banks Can't Transform Legacy Applications.

@JamesP: I can hardly think of a single bank on a mainframe core banking system in India but, in my personal experience, their uptimes are not any higher than those of any bank systems I've experienced in UK. Banking system failures of the nature visible to the outside world can be caused by switches, network, bandwidth, and many other moving parts. While legacy presents its own problems, I'm not sure if it can be singularly held responsible for all failures. 

James Piggot - Finastra - London 21 August, 2015, 09:07 0 likes

@Ketharaman for the Co-op bank their executives were probably talking about Why? for years which is "why" they are now in a real mess as reported here in Finextra today

https://www.finextra.com/news/fullstory.aspx?newsitemid=27756&utm_medium=DailyNewsletter&utm_source=2015-8-21

Ketharaman Swaminathan - GTM360 Marketing Solutions - Pune 21 August, 2015, 10:44 0 likes

Well, the last time Coop stopped asking "Why" and moved forward with legacy transformation, it almost went bust: "...calamitous IT re-platforming project that contributed to a £300 million write-off and left the bank on the brink of collapse." 

https://www.finextra.com/news/fullstory.aspx?newsitemid=26019

Unfortunately, Coop is damned if it transforms legacy and damned if it doesn't transform legacy. Its plight is more reflective of the way in which attempted to get out of legacy earlier and the way in which it is now attempting to stay with legacy, rather than legacy itself.

On a side note, I just came across a McKinsey article which prescribes two different ways of doing legacy transformation. Maybe the "two-speed" approach will suit most banks in developed nations and the "greenfield" approach will work better in emerging markets.

http://www.mckinsey.com/Insights/Business_Technology/Two_ways_to_modernize_IT_systems_for_the_digital_era

A Finextra member 21 August, 2015, 19:48 0 likes

Banks must not rush into transformation as an anti-legacy strategy. The foremost is to have a sound risk based information architecture, map it to systems. This will be one of the important criteria to make deciions. My view is replace because it is legacy is perhaps not the best way to make decisions. Business risk must be defined and constantly revisited. Coop is a great case study.   

A Finextra member 23 August, 2015, 21:55 0 likes

I see transformation happening through alternative means, rather than what is the accepted process of simply replacing a "legacy" system with a slightly new, slightly less legacy vendor platform. What truly modern core banking platforms exist? Many that purport to be are at least 10-15 years old in origin, and those are the ones that for the most part have seen least adoption. I really think the world is moving on from the concept of a "core banking system" as most vendors of such systems currently position them, towards "something" simpler. Maybe the establishment of new banks by the large banking groups will be the way forward...build a new bank with a new operating model and IT platform, and migrate customers to the new bank, allowing complete business transformation, not just IT platform. I disagree that banks on supposed "legacy" mainframe platforms will replace them because they "have" to...the tenure and compensation of CIOs etc is simply not structured in a way that makes this feasible, far easier to kick it into the long grass. Given the pace of change in fintech and the emergence of possible gamechangers, who would take the chance to sign-off a 15-25 year old supposedly "modern" core banking system?

Ketharaman Swaminathan - GTM360 Marketing Solutions - Pune 24 August, 2015, 08:47 0 likes

@DavidG:

You've made a brilliant point about how open CBS platforms are themselves not modern. Having worked for a vendor of one such platform, their average age is 20+ years.

Now that you've brought this up, I'm reminded of how the CIO of a leading bank recently lamented that even the mindset of executives in their open CBS vendor was tending towards legacy, with no sensitivity to agile development, time to market considerations in a crowded market, and so on. The bank wanted to build an instant account opening extension on top of the Online Banking module of their open CBS. The vendor took one month to come up the estimate, which was for $M / 6 months, which totally defeated the purpose of a functionality that had competitive advantage only for a couple of months.

The bank did something similar to what you've recommended, namely, contract with a third-party digital-first web development agency, launch the first version of the instant account opening portal in 30 days.

João Bohner - Independent Consultant - Carapicuiba 25 August, 2015, 12:56 0 likes

Why?

Because of the high 'inefficiency' of current systems - extremely high entropy.

The costs, only for maintenance of the systems are stratospheric compared to other industries.

It is necessary to modify the architecture of processes - and consequently the systems.

The processes must be performed 'corporately' instead of 'by-line-of-business' where redundancies take place, among other inefficiencies.

joao.bohner@gmail.com

A Finextra member 26 August, 2015, 13:05 0 likes

If the RoI for core banking transformation was solid, banks would be doing it left, right, and centre surely? Either the economics vs risk do not stack up to make it sufficiently compelling, or core banking vendors have not done a good enough job in their value propositions to help the banks build a compelling business case.

What banks are doing is spending on the stuff that is highly visible to their customers, predominantly mobile/omni-channel/digitisation efforts. Plainly they see a better business case for doing that.

Whilst some UK banks have had very public meltdowns of back office technology that has caused massive disruption to their customers, if you look deeper at the cause very often it is operational and process related, rather than specific issues with the technology in use. It's easy for self-interested core banking vendors to simply wheel out phrases like "creaking infrastructure" and "legacy technology" every time this happens.

Ketharaman Swaminathan - GTM360 Marketing Solutions - Pune 26 August, 2015, 14:19 0 likes

@DavidG:

I couldn't agree with you more:

As I'd highlighted in Why Banks Can't Transform Legacy Applications - Part 2, bank C-Suite members, who run the most profitable industry in the world, get cheesed off when revenue-challenged, loss-making, VC-funded fintech upstarts preach to them about how to cut costs by transforming legacy!

Applying the wrong patch, failure to test the new release - operational and environmental factors like these that have led to most UK bank system failures. They could happen equally well with open systems as with mainframes.