Conquer your Systems of Record
It is quite entertaining to see the many bloggers climbing over each other trying to come up with the “best” top 10 predictions for 2016, or list what’s hot or not.
Well, I’m going to go against the stream and instead start 2016 with a post about an older topic — getting teams and legacy systems that support core transaction processing and manage critical data to remain relevant in a fast changing, real-time, 24/7 and always connected society.
I’m already hearing “That’s sooooo yesterday’s news…” but the fact remains that despite billions of pounds invested in IT improvements and resilience programmes organisations continue to struggle. There have been several high profile system failures in recent years such as the RBS software issue in 2012, for which they were later fined £56m, but you had HSBC’s online banking customers suffering problems as recently as last week. It’s not only banks that have problems with their IT systems, but it tends to be what you hear about in the press.
I would be naïve to try and pen down an answer in this blog to the many, and often complex, challenges organisations face, but I want to highlight a few areas where improvements most certainly can be made.
You have probably heard about Systems of Engagement, a definition that Geoffrey Moore, coined years ago in his 2011 paper Systems of Engagement and The Future of Enterprise IT. He uses Systems of Engagement to refer to the transition from legacy systems designed around discrete pieces of information (“records”) to systems which are more decentralised, incorporating technologies that encourage collaboration and connecting people in real time. His view is that this transformation will be enabled by new Systems of Engagement that will overlay and complement the deep investments that have already been done in an organisation’s Systems of Record.
Many organisations have gone down this path, but unfortunately focusing on the new systems while the old ones are left to wither away under “business as usual”. The reality is that unless the focus shifts from the latest tools and technologies to one that also covers people, process and the ageing systems, the gap between teams and technology generations will become too wide to bridge.
Gartner published their Pace-Layered Application Strategy a few years back as a methodology for categorising applications and developing differentiated management and governance processes. While a bit dated, it can still be useful for highlighting different aspects of an organisation’s application landscape. Gartner’s strategy also introduces something called Systems of Innovation, which are new applications that are built on an ad-hoc basis to address emerging business requirements, typically with very short project life cycles.
The reason I mention Systems of Innovation is that they sit at the other end of the spectrum from Systems of Record when it comes to agility and rigour. Systems of Innovation are also what tends to excite business executives who, once getting accustomed to very fast delivery times, get frustrated with the slower pace of change within Systems of Engagement and particularly Systems of Record.
Systems of Innovation don’t necessarily integrate with backend systems and often stay within Development or Test environments, never reaching Production, which further distorts the picture of certain teams being able to deliver applications in days while others seem to struggle as they need months, if not years, to deliver.
While organisations strive to get more agile and move up the pace layers, many large organisations have no other option than continuing to depend on their old Systems of Record, at least in the medium term. The effort involved to phase them out is high and will often take years to complete so organisations need to find ways to extend and leverage what they have, or risk being overtaken by smaller and nimbler competitors. The flip-side is that these systems, and the information within them, have been built up over years, if not decades, so they shouldn’t be undervalued as an asset when it comes to fending off the competition.
Organisations may get this, but fail to recognise that things like Agile and DevOps aren’t going to deliver the same outcomes when it comes to Systems of Record. Investment is also needed, not so much in technology, but in people and processes. Unfortunately, the opposite often happens with budget cuts and talent moving on, leaving understaffed teams with skill gaps in charge of operations, which undoubtedly have a negative impact on things like system resilience. If organisations are to successfully leverage the inherited value in their Systems of Record, it’s time that they take a step back and re-evaluate how they deal with these systems and the people maintaining them.
If you’re making plans for your legacy systems, don’t forget your staff
We all know that organisational change is part and parcel of any IT transformation programme, but the pace of change, with literally every organisation scrambling to jump on the digital bandwagon, has really shaken up IT departments in more recent years. Hailed as the future of IT only a few years ago, some teams find themselves in the “legacy bucket” and their platforms earmarked for decommission, often without having a say in the strategy that put them there.
The problem is that while it is easy to get excited about the promises of new technologies, the impact of introducing them is often underestimated. Not only does it end up taking longer time and creating new challenges, but it also require support from the very people that have just been told they are no longer “relevant”.
If plans are put in motion to migrate off a legacy system, it is important to engage its team from an early stage. Inevitably there will be resistance, but the success often stands and falls with their support. A recipe for disaster is to encourage people to leave (e.g. redundancy, retirement) because it may create a knowledge gap that will be very hard to fill if needs arise later on. The ideal scenario is to make the old team part of the new world, not write them off together with the legacy system.
Don’t do agile, be agile
Oh no, that old cliché. Well, as long as organisations continue to say that they want to implement Agile, or DevOps, seeming to believe it is a “thing” you do, it remains relevant. Agile and DevOps may require new tools & processes to be brought in, but it is a culture change that affects people more than anything else.
It’s also important to realise that you can “chip away” at one area at a time, slowly realising many of the sought after benefits over time. This is also relevant for systems and teams earmarked for decommissioning as it may take years before they cease operation.
Take something like the Agile principle to “Deliver working software frequently, from a couple of weeks to a couple of months, with a preference to the shorter timescale”.
It simply says that you should strive towards more frequent releases — this can be achieved for any system, even Systems of Record. New tools may help, but the focus should really be on changing the way people work and getting rid of unnecessary processes — ever wondered who read that 50-page release document that you were asked to print and sign before going live?
Many teams are fighting just to “keep the lights on” so be prepared to invest in them and the way they operate, particularly if you’re looking to eliminate time-consuming and error prone manual processes.
Automation doesn’t come for free, but there are some great benefits down the road
You may have come across continuous delivery, an automated software engineering approach in which teams produce software in short cycles, ensuring that the software can be reliably released at any time. The approach promises more efficient testing, streamlined workflows, and higher confidence in operations by allowing for more incremental updates to applications.
While few dispute the benefits of continuous delivery there are some practical challenges involved, particularly when looking at introducing it for old and complex legacy systems. Teams feel increasingly frustrated and powerless as they’re asked to automate without anyone wanting to hear about these challenges. The last thing they need are people throwing new tools at them.
Many legacy systems require unique skills, skills that often only an ageing and specialist workforce can provide. Teams may also not be up-to-speed with the latest trends and feel hesitant about change they don’t fully understand, so it is important to create a good and open dialogue to reach a mutual agreement on the way forward.
While true continuous delivery may be out of reach for many of these systems, incremental inroads can still be made as long as teams manage to secure the necessary funds, resources and executive support they’re often lacking. Benefits will not come overnight, but even small steps will help over time, even for Systems of Record.
Some final words
Nothing I’ve said in this post is “hot off the press” and it would probably have read very much the same if I had written it a year ago. The thing is that these days everyone seems to have something to say about the latest technologies while less and less is being said about the pressure and challenges facing many of our “traditional” IT engineers and the systems they work so hard to maintain.
It’s time that organisations wake up to the fact that Systems of Record will continue to provide much value over years to come despite not appearing in the press — at least not for the right reasons — so let’s start 2016 by giving them the backing they truly deserve!
Originally published at www.linkedin.com.