This is the text of a keynote I delivered at FinovateSpring 2026 in San Diego.
As most of you learned at breakfast, I’m John San Filippo, publisher of Finopotamus – Fintech News for Hungry CUs. I’ve been in the credit union technology space for 37 years. Back in the day, I was Symitar’s 54th employee. I worked for Symitar three different times, worked for Bluepoint Solutions, since purchased by Alogent. And somewhere in there I also started another magazine, Credit Union Business. So yeah, this ain’t my first rodeo.
Before we get started, I have a couple of important disclaimers. First, I don’t have a slide deck today. Some people are masterful at conveying their message via PowerPoint. I’m not one of them. I may dazzle you with my words of wisdom or I may put you to sleep, but either way, it’ll be all me.
Second, I’ll be pretty much glued to this podium. Again, some people look very natural and relaxed roaming around the stage. Again, I’m not one of them. For better or worse, I have two titanium knees that I don’t fully trust and I would rather not end up in somebody’s lap.
In the Olden Days
So in my nearly 40 years in this space, there was one immutable rule that seemed almost chiseled in stone. If you hit a brick wall with your current core platform in terms of feature/functionality – or service, for that matter – you had two options. Option 1: Become a master of work-arounds and make do with what you have. I’ve seen people do it. I wouldn’t recommend it. Or Option 2: Go out shopping for a new core to replace the old one.
We all know Option 2 is easier said than done. We’ve heard – or maybe even lived – the horror stories. In fact, I’ve heard a core conversion likened to performing open-heart surgery on a patient as they’re headed for the finish line at the Cherry Blossom 10 miler.
I have some good news, my friends. As you embark on your quest for core modernization, boldly marching into our brave, new, API-first world, there is now a third option. It’s called a sidecar strategy. Or you might hear it as a speedboat strategy. Same thing.
What Is a Sidecar Strategy?
How many people are familiar with the term sidecar as a core modernization strategy.
Incidentally, in terms of metaphors, I prefer speedboat. I mean, think about it. I sidecar can only go as fast as the motorcycle it’s connected to. On the other hand, a speedboat can go … well … pretty damned fast, all on its own. But whatever. Sidecar is the more common term, so we’ll stick with that today.
So what is a sidecar strategy? Instead of finding a new core to replace your old core, you find a new core to add to your old core. Your heard me right. You voluntarily elect to run two cores in tandem.
I know what you’re thinking. John, why on God’s green earth would I possibly want to run two cores? It’s hard enough just running the one.
The answer boils down to speed to market, because in the world today, it’s not only what you offer; it’s also how quickly you can offer it.
Following Fast Isn’t Fast Enough
For years, I’ve heard credit unions almost brag about being so-called fast followers. I’ll keep my eye on this Internet banking thing for a couple of years and if it catches on, I’m all over it. Well, I have some bad news, Skippy. Today you don’t have two years to wait and see. You might not even have two months to wait and see. So go ahead and be a fast follower. Just promise me you’ll adjust your definition of fast down to a matter of weeks or even days.
That brings us to the first major advantage of a sidecar core. It’s not the functionality it brings to your credit union; it’s the functionality it brings to your credit union now. Think about it. How long does a core search take? Then if you’re looking at a popular core, your conversion date might be a couple of years out. From the time you say, “Holy crap, we need a new core,” to the time you actually have one can easily be three years or more. What happens when you’re forced to wait three years to innovate? I can almost promise that the world will pass you by as you’re stuck in that pre-conversion holding pattern. You simply do not have that kind of time to waste.
On the other hand, tech providers that specialize in this area can stand up a sidecar core in a matter of weeks if pressed to do so. Since you’re keeping your old core, your employees will continue to work as they always have. In the meantime, you have a powerful new platform with which to innovate, iterate, collaborate and get as crazy as you want – all with very little long-term risk.
How About Some Use Cases
What are some of the most popular use cases for a sidecar strategy? At the top of the list is the ability to power a niche banking initiative with no disruption to your existing brand or operation. I’m talking about creating a discreet digital brand that caters to some market niche. You’ve seen those digital-only banks that cater to gig workers or doctors or freelancers. That could be you. Or when you consider that 30% of retail credit union accounts are really small business accounts in disguise, maybe it makes sense to give those businesses their own dedicated brand with their own dedicated business tools.
Even if you decide you want to support businesses through your primary brand, your legacy core may not have all the features you need to do so in 2026. A sidecar core with good commercial banking chops can take care of that for you.
We all know there’s real real time and fake real time. Any core can handle fake real time, but when it comes to real real time, not every legacy core is ready to face the challenge. Payments are a good example. I imagine just about any core could handle a fake real-time payments scheme like Zelle. But when you get into real real-time payments like FedNow or RTP, a sidecar core may be able to handle the job more efficiently.
Then there’s the potentially big money maker: banking as a service, or BaaS. All those B2C fintechs out there need some sort of real financial institution at the back end to handle all the banky stuff. Do you know who the sponsor bank for CashApp is? It’s Sutton Bank. Never heard of them. How about Affirm? Ever hear of Cross River Bank. Me neither. My point here is, you don’t need to be a Chase or a Citi or a Wells to be a sponsor institution for a fintech. You just need a valid charter and the right tech stack. With those two things, the sky is truly the limit.
Devil’s Advocate
There are plenty of core providers out there who will gladly offer you more detail on the benefits of a sidecar strategy. Mambu comes to mind. Nymbus comes to mind. Even Jack Henry and Fiserv come to mind. I’m sure there are others. Since these companies have the cheerleading squad covered, I’ll mix a metaphor and serve as devil’s advocate for a moment. What are some of the challenges of a sidecar strategy?
The obvious one is cost. If you spend X running your current core, you’re going to spend X plus Y running two cores at the same time. Is it worth it? Of course you need to do all the ROI math. Compare your TCO to the projected revenue from your sidecar-based initiatives. Crunch the numbers. Etcetera, etcetera, etcetera.
However, you also need to keep in mind that there’s one thing it’s impossible to measure. That’s the value of the member who didn’t move to the bank down the street because you were able to offer the right product at the right time. Put another way, you can’t measure something that didn’t happen. And with the right sidecar strategy, there should be a lot of member defections not happening.
That’s something I learned from Jack Henry – I’m talking about the man, not the company – and it’s something credit unions know better than anybody – if you focus on doing what’s right for the customer/member, the money will follow. It always does. I’ve never seen that not be true.
Let’s not overlook the investment in time and talent that’s required to pull off something like this. Just like it takes more money to run two cores, it takes more people to run two cores. Or does it?
Remember how we’ve been telling our employees that AI won’t replace them – how AI will only relieve them of their more mundane tasks so they can focus on more strategic initiatives? I can’t think of many initiatives that are more strategic than successfully deploying a sidecar strategy. This could be the perfect confluence of AI providing you with more available humans at the exact moment you actually need more humans in the loop for a very specific strategic initiative. Play your AI cards right and you could get through this whole thing without adding a single FTE.
Big Data Is the Big Problem
Of course, by far the biggest challenge in deploying a sidecar core is getting a handle on all the data. It was probably 20 years ago when I first heard a conference presenter telling credit unions about the treasure trove of data that was hidden away in their cores and how it was being grossly under-utilized. Quite frankly, I’m stunned at how little progress has been made in this area in the ensuing two decades. Credit unions do recognize that they have this valuable data at their fingertips, but they still struggle with what to do with it. Now factor in data coming from a second core and you have the potential for a real cluster.
But it’s not just a matter of keeping your data organized; it needs to stay synchronized. After all, in the end there can be only one version of the truth. That’s Data Analytics 101. If the sidecar says the member has $500 and the legacy core says $450 because of some batch delay, which one wins? If you don’t have an immediate answer, you don’t have a strategy; you have a lawsuit waiting to happen.
Data management really is the single most important component in any successful sidecar deployment. Get it right and you’re off to the races. Get it wrong and you’re in a ditch on the side of the road spinning your wheels in the mud.
If you’re already in a good place with your data, you have a great foundation for sidecar success. If you’re struggling with your data, you need to fix that now so that you’re ready when your sidecar gets introduced into the mix. And don’t be afraid to ask for a little help from your friends. Your sidecar provider has special expertise dealing with data issues as they relate to this type of project. Likewise, there are plenty of great data consultants out there in Credit Union Land. And the one thing all these potential tech partners have in common is that they want to see you succeed.
Planning for the Future
The last point I want to drive home is that a sidecar strategy is intended to address a current need. It’s not intended to be a permanent fix for anything. As your operation becomes more and more sophisticated, you’re going to want less and less to be juggling two cores. I guess what I’m saying is, when you go shopping for a new sidecar partner, you’re really shopping for your next core platform.
It’s just that unlike the days of yore, you can ease into your new core on your schedule, without a nasty traditional core conversion, while still taking advantage right now of all the feature functionality that core has to offer.
Is “rip and replace” officially dead? It’s probably too soon to ring the death knell just yet. But does “rip and replace” need to make room for sidecar as a viable core modernization option? Absolutely.
Thanks for your time.