Data modernisation and digital transformation projects are no easy tasks for businesses to undertake. Little wonder, then, that it’s often treated as the last thing businesses want to do. Done badly, they can be disruptive and unruly, with significant ramifications for an organisation’s way of working.
“Modernisation, in the past, has sometimes been seen as a plumbing project,” says Thilo Rockmann, CEO of LzLabs. “Everyone’s got an ‘if it isn’t broke, don’t fix it’ mindset. And that’s a real issue because often these applications aren’t technically broken at the point they need modernising.”
Another problem is that initial projects often tackle small, simple areas of a business, with leaders presuming that biting off less of the problem will make it easier. But if one area is modernised and updated and the rest of the business isn’t, it can cause issues of incompatibility and broken dependencies among a complex set of applications and systems.
There’s often a sense of having renovated one area of a property, only to find doing so has uncovered a small mountain of issues with the rest of the building. “There’s a huge learning curve in there. People have often hit that brick wall headfirst without a hammer, unfortunately, because digital transformation is just as much a people process as it is a technology process,” says Rockmann.
Why thorough modernisation is important
Although it may seem like a headache, modernising legacy technology is a vital requirement for businesses. Experienced developers are becoming more scarce, younger employees are unlikely to be familiar with older applications and the talent pipeline is not replenishing institutional knowledge within organisations. That’s clearly problematic if organisations are relying on old-fashioned systems that underpin their core business.
This is not a one-time exercise but a continuous and recurring process
“You have a small percentage of people who can deal with these old languages, and then a percentage of those that can deal with the old technologies, and then a percentage of those who have institutional knowledge of the applications,” says Rockmann, “so you’ve got fractions times fractions times fractions, and it gets to be a scarily small number of people at the end.” It’s a ticking time bomb that could explode at any minute. If anything were to go wrong with a long-established application, the fix may not be particularly complex but it might rely on a very specific bit of knowledge. If there’s no one remaining at the company able to make that repair, it could leave a business in a sticky situation.
Modernisation helps keep IT and business systems up to date, meaning that they can be agile and flexible to business needs – not just whoever has the long-term memory of when it was first installed. Naturally, a need to keep systems up to date means that modernisation is never a one-and-done project, explains Rockmann. “This is not a one-time exercise but a continuous and recurring process,” he says. “It’s a process that can be approached in a variety of ways.”
Moving beyond simple solutions
Rockmann points out that prior history has recommended following the principle of the seven Rs when considering how to modernise applications. The majority of companies on the cusp of transformations expect to replatform or replace, though other approaches such as rehosting and refactoring are gaining in popularity.
The seven Rs
The basic approaches to application modernisation
Replatform: Recompiling an application on a new platform or operating system.
Rearchitect: Changing how the systems work, often splitting them up into smaller, modular forms.
Rebuild: Starting from scratch and building again – this can be costly and risky but leads to a clean, modern deliverable.
Rehost: Shifting from one environment to another, most commonly associated with moving wholesale to the cloud, with no changes to language or source required.
Refactor: Changing code within a system without changing functionality to adapt to new advances. This is often achieved by transcoding from older languages with automation.
Replace: Swapping out existing systems for a fresh set of applications, bought especially for the project. Most commonly off-the-shelf packages for non-differential applications.
Retire: Giving up the ghost – or the old way of working – and potentially (but not always) replacing it with alternatives which may take different forms.
However, leaders can fall into the trap of simplifying a modernisation project by picking one approach. Leaders shouldn’t feel like they have to commit to just one of the seven Rs, as even approaches like replatforming and refactoring won’t solve all the issues. For example, refactoring one element of an organisation’s IT infrastructure will likely identify the need to take different actions elsewhere. In some cases, replacing one of the Rs with two of them sequentially can unlock value quickly and reduce risk – often taking problem issues out of the critical path. “The seven Rs aren’t a bad approach, but they’re not elegant when one approach is applied to an entire portfolio,” says Rockmann.
In part, that’s because the seven Rs are designed to solve simple problems – and as any business leader knows, that’s not the reality of any organisation today. Technologies to assist transformations have also evolved beyond these simple approaches. “Modernisation 1.0 is doing the easy stuff, without the dependencies,” says Rockmann. “And because they’re learning, they’ve actually found some of that quite hard still. Now, they look internally and see the Gordian knot, and this is the ‘Oh no’ moment. That simplistic, single approach won’t work.”
LzLabs’s experience of helping companies enact digital transformations successfully shows that there are no simple big bang fixes. “We believe in an iterative, incremental and interoperable approach,” says Rockmann. “We don’t believe that there is a one-size-fits-all solution. Not even ours alone.”
Business leaders need to indeed look at their systems holistically and find a bespoke solution. They also need to come to terms with the fact that almost any modernisation project will also raise further issues beyond those first identified. Being cognisant of that – and comfortable that it’s an ongoing, multi-step approach, rather than a quick project that races to completion – is crucial to tackling the Everest of a full modernisation project.
The trick is not to try and tackle the full climb in one go, leaving leaders overwhelmed and organisations stuck at base camp. A step-by-step approach broken down into small stages is the key to reaching the cloudy summit and making worthwhile change within the organisation quickly, with low risk. Keeping front of mind one thing is vital, explains the LzLabs CEO. “Modernisation is doable,” says Rockmann. “But you need to think in a different way.”
For more information please visit lzlabs.com
Data modernisation and digital transformation projects are no easy tasks for businesses to undertake. Little wonder, then, that it’s often treated as the last thing businesses want to do. Done badly, they can be disruptive and unruly, with significant ramifications for an organisation’s way of working.
“Modernisation, in the past, has sometimes been seen as a plumbing project,” says Thilo Rockmann, CEO of LzLabs. “Everyone’s got an ‘if it isn’t broke, don’t fix it’ mindset. And that’s a real issue because often these applications aren’t technically broken at the point they need modernising.”
Another problem is that initial projects often tackle small, simple areas of a business, with leaders presuming that biting off less of the problem will make it easier. But if one area is modernised and updated and the rest of the business isn’t, it can cause issues of incompatibility and broken dependencies among a complex set of applications and systems.
