AI at Congrify: Building a Payments Data Co-Pilot
By Marco Conte
The Spark That Started It
When ChatGPT burst onto the scene, it flipped a switch for me. I have spent 11 years working in the payments industry, including some years building risk & fraud intelligence leveraging supervised and unsupervised machine learning models, but this was next-level: AI that could talk back, understand you. I thought, “Payments need this.” Not cold algorithms, but a conversational ally. With the progress in LLMs and Agents, it became clear that AI in payments could transform how merchants manage their payment operations. We envisioned an AI agent tailored for the payments industry, built on top of our data infrastructure, to deliver hidden insights and automate operational processes faster than ever.
That vision crystallized when I met a Head of Payments at a European merchant during an event in Berlin, Germany. He was testing the first prototype of Congrify; it was still buggy with a poor design and user interface, but he was impressed by the speed of querying the data. He said, “I imagine starting my day with a unified payments portal, grabbing a coffee, and in minutes, having all the payments reporting I need to see how the business is doing.” Though he didn’t end up using Congrify at the time, he was amazed by the payment analytics—getting all his data in under 20 seconds.
Now, we’re building a constellation of AI agents that deliver even more: insights that feel like magic, before your coffee’s cold.
The Problem We Couldn’t Ignore
A decade ago, I witnessed firsthand the struggles merchants faced—a chaotic tangle of inefficiencies that mirrored my own experiences at Western Union. Back then, the problem was clear: too many tools and data sources, not enough actionable answers. Fast forward to today, and shockingly, little has changed.
Payment portals continue to multiply, dashboards grow increasingly convoluted, and clarity remains elusive. Merchants are forced to spend countless hours piecing together fragmented data from various payment service providers, just to identify a sudden fee spike, understand troubling payment declines, or uncover trends that could shape their strategies. Hours upon hours are wasted—time that should be invested in scaling their businesses, nurturing customer relationships, or innovating new offerings.
This fragmentation isn’t just a minor inconvenience; it’s a pervasive issue that drains resources, erodes profits, and chips away at merchants’ sanity. The financial toll is undeniable, with inefficiencies leading to missed opportunities and hidden costs that accumulate over time. We knew this couldn’t continue. Determined to bring order to the chaos, we turned to AI and data science, harnessing their power to transform raw, disjointed data into meaningful payments intelligence that empowers merchants to make informed decisions with confidence and speed.
Our Vision: A Payments Partner
I always imagined saying, “Now you have a new buddy for your payments management.” We thought of many names, but simplicity won. It’s a buddy that inspires trust and helps with daily tasks—quickly. We know everyone’s busy, but being busy doesn’t help merchants unlock their payments’ full potential. Too often, companies unknowingly leave money on the table, or payment professionals drown in dashboards and manual analysis.
That’s exactly what we’re changing. With AI agents, hours spent on routine tasks can be reclaimed for work that truly matters—strategic decisions, creative problem-solving, and meaningful collaboration. It’s not just about saving time; it’s about empowering teams with payments intelligence to make smarter, faster choices.
Payments knowledge in the industry can be very hard to find, this is a very complex combination of technology, finance, pure payments domain knowledge. Bringing an AI buddy for payments should help anyone, new or not to this industry to learn more in understanding how to master payments.
Why AI? Why Now?
AI excels in environments where repetition and complexity intersect, automating tedious tasks and unraveling intricate patterns that humans might miss—but its effectiveness hinges entirely on the quality of the data it processes. Without clean, reliable data, you’re stuck with the age-old problem of ‘garbage in = garbage out,’ where flawed inputs lead to meaningless outputs.
At Congrify, we recognized this challenge early on and dedicated years to meticulously engineering robust data pipelines that ensure accuracy and consistency. Our efforts mean that the data feeding our AI systems is as dependable as it gets, laying a solid foundation for transformative insights. We’ve paired these pipelines with advanced machine learning models that can detect anomalies—like unexpected spikes in costs such as interchange fees or unusual transaction patterns—saving merchants from costly surprises.
On top of that, we’ve integrated cutting-edge LLMs to translate these findings into plain, conversational English, making complex data accessible to everyone, not just data scientists. This combination is nothing short of a breakthrough, offering merchants a seamless way to understand their payments landscape without getting bogged down in technical jargon.
So, why is this happening now? The answer is simple: merchants are hemorrhaging both time and revenue due to outdated, manual processes that can’t keep up with the demands of modern commerce. These inefficiencies create a ripple effect—delaying decisions, obscuring opportunities, and ultimately stifling growth. AI in payments is poised to change that narrative, delivering real-time, actionable insights that empower merchants to take control of their financial operations. The technology is ready, the need is urgent, and the time to act is today.
Why This Matters to Me
We’re a team of people who’ve spent years in the payments world. Personally, this is my 12th year in the industry, and I still feel genuinely excited every day. Payments is one of those rare spaces that’s constantly evolving, never boring, and endlessly impactful. I feel lucky to be part of it.
But I’ve also lived the frustration firsthand—spending hours digging through spreadsheets and dashboards, chasing answers that always seem just out of reach. It’s exhausting, and it pulls focus away from the work that really matters.
That’s why we don’t just build for our customers—we partner with them. Every day, we support payment teams facing complex data challenges, and we take that responsibility seriously. We know how much smoother and more meaningful their work can be with the right tools in place.
And honestly? That’s the most rewarding part—seeing the look on someone’s face when they realize how much time, energy, and stress they can save with a tool that just gets it.
Sure, even AI can’t fix bad coffee—but it can absolutely fix bad payment processes.
And that’s a pretty great start.
What’s Possible
Imagine opening one portal, asking, “How’s my business doing?” and getting a full rundown—payment declines, interchange fees, scheme fees, provider fees, chargeback rates and other trends—in under 20 seconds. No bugs, no fuss, just insights that’d take hours otherwise. That’s the AI Co-Pilot we’re building, a complementary part of our payment analytics and instant payments reporting capabilities. Want a taste of what it does? Our first blog post dives into its magic—give it a read.
Ready to see it for yourself? Bring your coffee, and let’s talk payments!