In the whirlwind world of commerce, staying one step ahead is the name of the game. But there are still many brands that are straddling old-school mainframe tech. It’s like trying to run a marathon in flip-flops! Upgrading to the latest tech architecture can feel like a giant leap, but the agility and business outcomes it yields are worth it. Let’s dive into how we can make this jump with as few hiccups as possible, focusing on three major points and sharing some real-life stories along the way.
Taking a Good, Hard Look at What We’ve Got
The Challenge: Ever tried untangling a bowl of spaghetti? That’s what it feels like moving away from mainframes. Not only are you battling typically decades of legacy code, but there are often ‘renegade’ jobs that are running that, if ‘unplugged’ may inadvertently wreak havoc on your daily operations.
Getting Through It: First up, put your detective hat on. We need to map out where and how these mainframes are orchestrating and processing data interchange between the broader landscape. Think of it like planning a road trip – you need a good map of the customer’s journey (internal and external customers) and understand what the technology is doing to ‘service’ those flows! Break it down step by step, focusing on where your hotspots are in the journey, then clearly seeing where the underlying technology is that may be causing the gaps.
Picture a big retail chain realizing that everything from stocking shelves to ringing up sales is tied to their mainframe. They decided to first spruce up their customer service systems with some cloud magic, keeping the shopping experience smooth while they worked on the back-end systems. This is typically not the right approach because it’s putting a better interface on a set of systems that are not real-time.
Rapid Value-Delivery, but with a “Agile-Fall” Approach
The Challenge: Likely you have a program management office that is not fully comfortable with just “going for it” in a pure Agile way. We see the rise of frameworks like SAFe and other approaches that let you approach things in planning increments but also allow intra-sprint flexibility in achieving the outcomes (through key results) that you want…. but, the issue remains, how do you DE-construct a mainframe solution in a composable way, but still have a sense of projected cost, staffing and the ability to manage / navigate the risk.
Getting Through It: Largely, the existing mainframe is a mix of a pseudo CRM, the back-office finance functions, and some degree of customer-service-meets-logistics-enabler (i.e. “OMS-light”). The good news here is, you have cloud-native solutions now that you can rapidly spot implement as you interate through the decomposition of the mainframe. Think about it in terms of one capability group in the above map you created. One capability is order routing and order orchestration, which, is typically one of the highest value drivers of a move to a modern OMS. You can now implement just that function from an OMS and have the mainframe still serve as “main” orchestrator but to punch out to the API’s that manage order-routing. NET – You now have much more choice in how you progress through the decomposition.
Moving Data Without Dropping It
The Challenge: At one point we’ve all moved apartments/houses after living in one place for years. That’s what data migration feels like. It’s crucial to move this data safely without losing or jumbling it up. Data migration strategies must be documented, articulated and ‘dry run’ as you progress your infrastructure from mainframe to modern ERP + OMS.
Getting Through It: Let’s sort our data into “must-haves” and “nice-to-haves.” We move the important stuff first and make sure it’s squeaky clean and secure. Think of it as packing your valuables with extra bubble wrap. If you are currently storing what is effectively a lightweight CRM in your Mainframe, you’ll need to consider moving that to a commercial SaaS / Cloud solution and planning for doing this in a composable way as part of a broader IT strategy.
Teaching Old Dogs New Tech Tricks
The Challenge: Swapping to new tech often means learning new skills. Some of your existing team may need to be re-skilled, while others may be at or past the point of wanting to retire, so you’re likely facing a scenario where you must extract all the knowledge of a very seasoned employee on the team while simultaneously navigating the KTLO (keep the lights on) work. This is where a staffing strategy for the project and an operating model for the new landscape comes into light.
Getting Through It: It’s time to bring in new perspective and talent, and allowing them the time and space to integrate, learn and accelerate the transformation! You’ll want to create a culture of rapid experimentation and failure, but also ensuring the PMO and executive leadership teams know the budget and timelines are going to be what you said they were. If you have a stakeholder that is a bit nervous about ditching their mainframe, take the time to understand those concerns, address them within the strategic framework on how you’re approaching the transition, and ensure their team is brought along for the ride (NOTE – I often see this coming from the CFO, and rightfully so).
How to Get Started
Transforming your tech from legacy mainframes to the latest and greatest doesn’t have to be a nightmare. It’s all about smart planning, careful data handling, and empowering our teams. Remember, there’s no one-size-fits-all solution. Tailor these steps to fit your unique culture, viewpoint and capability, and you’ll be setting your business up for a future filled with agility and growth. If you want help thinking through your approach, reach out to us below and let’s start a conversation.
