Introduction: Meet Julien

Julien started his career in the mid-2000s, when DVDs were still around and streaming services were unthinkable. Over time, he became fascinated by the rise of smartphones and how technology reshaped traditional industries. Eventually, his passion for problem-solving led him to co-found Agilytic, a data and AI practice that helps organizations solve real business problems. Whether it’s reducing late payments, making sense of customer data, or tailoring AI to an organization’s distinct challenges, the goal is always the same: make data useful.

In the latest episode of our podcast, Julien explains why data is much more than a buzzword, how companies can put data at the center of strategic decisions, and why “data hygiene” matters more than catching the newest AI hype-train. He also shares practical tips for small and medium-sized businesses on how to get started, and why the best data projects always begin with clear, well-defined problems.

Q&A: Data, AI, and the power of combining information sources

Julien, what initially drew you to data and AI solutions?

I’ve always loved solving problems in a factual, evidence-based way. Early in my career, I worked in telecom, focusing on customer retention. I realized how powerful it was to rely on data, not on personal opinions. Data made my job convincing executives much easier: instead of “Trust me, I’m an expert,” it became “Let’s test this hypothesis with real numbers.” That was a turning point. It’s still the foundation of Agilytic: focus on the real problem, rely on the facts, and get tangible results people can trust.

Why do you emphasize “problem first” over “tech first”?

Because falling in love with a shiny solution, such as a chatbot or an AI-powered interface, often ignores what you truly need. We hear companies say, “We need ChatGPT to boost our business,” when their real issue is poor data quality or inconsistent customer records. You can deploy the fanciest, most complex tool, but if your underlying data hygiene isn’t solid, you’re just automating the wrong processes.

What matters is: What problem are you solving? Are you trying to reduce churn? Speed up order fulfillment? Identify high-risk clients? If you figure that out first, then you can pick the right combination of data pipelines, analytics, and AI. Sometimes, a simple rule-based system is enough. Other times, it’s advanced automation.

Speaking of data hygiene, why is it so important?

Good data hygiene is the backbone of any successful data or AI initiative. If you want to create that single repository of good information, what I call an organization’s grey matter, your data must be well-structured, accurate, and accessible. Otherwise, it’s garbage in, garbage out.

For instance, many businesses ignore internal data sources like sales invoices or shipping logs. Yet invoice data alone can reveal your most profitable clients, which regions generate the most revenue, and how often customers miss payments. From there, you might see patterns that help optimize marketing budgets or develop advanced AI models. Without that foundation, you can end up chasing sensational “solutions” that never solve anything long term.

Can you share a real-life example of data-driven problem-solving at Agilytic?

One of my favorites involves late payments. We worked with an insurance company whose customers weren’t paying premiums on time, creating a cash flow nightmare. Many organizations jump straight to sending threatening letters, but we discovered that “one size fits all” was the bigger problem.

We analyzed who was late and why. It turned out 60% (hypothetical figure) were just forgetful. They weren’t bad customers; they just missed deadlines. Another group felt frustrated because they had outstanding claims that hadn’t been settled. A smaller segment had no intention of paying.

By segmenting those profiles, the company tested different approaches for each group: friendly reminders for the “distracted” customers, deeper support for clients waiting on reimbursements, and firmer action for the truly overdue. They ran these tests on carefully selected samples, measured the outcomes, and found that customizing the approach dramatically cut unpaid bills. Ultimately, it saved them a lot of headaches, and a lot of money.

How about a B2B example? People often think “not enough data” for B2B.

Absolutely. Many B2B companies feel they have too few clients to do “big data.” But B2B data can be incredibly rich. Take a manufacturer that sells to businesses across multiple regions. Each client has a company registration number (in many countries), which unlocks a host of firmographic data: annual revenue, industry, location, growth over time, you name it.

You can also add external information, like local demographics or energy prices, and overlay everything on a map, this is where geospatial analysis shines. By stacking data layers (like your internal sales logs and external market intelligence), you see hidden opportunities. You identify which areas to target, who might respond to a certain message, or which addresses have the greatest potential for solar panel installations. You don’t need a million clients; you just need the right data.

Do these data solutions help smaller companies too? Where should a small or medium business start?

Yes, and the trick is to avoid thinking, “I’m too small to benefit from data.” Even a small family business has a billing system, a customer database, or a CRM, though many CRMs are underused. Start by inventorying your existing data. Ask questions like:

• What financial and sales data do I have? These are often the best starting point, since they're tracked more carefully.

• Are my customer records rich or deep enough to segment by behavior or location?

• Can I tap into open data (for example, local market info) to enrich what I already have?

That’s your foundation. Then, define what you really want to solve; maybe boosting customer loyalty or streamlining your stock levels. Once you know the problem, you can see if your existing data suffices or if you need more. From there, it’s easier to spot public data sources that will greatly deepen your insights. Layering a variety of both internal and public data sources over each other feels like having a superpower. Smaller organizations are often amazed at how quickly they get results with the right focus.

What’s your advice to companies who say, “We need advanced AI or we’ll miss the boat”?

AI is powerful, but it’s not always the answer. Sometimes, you just need strong dashboards, clear sales reports, or a basic rules-based approach. There’s no point spending big money if a simpler fix meets 80% of your needs.

Part of Agilytic’s commitment is an authentic desire to help. We’ll say “no” to a big AI project if you can fix the issue with simpler, cost-effective steps. That honesty builds trust and prevents the “solution in search of a problem” scenario.

Bringing It All Together

Many lessons emerge from our conversation with Julien, and they center on pragmatic, people-focused problem-solving. Data and AI thrive when they address concrete challenges: like cutting overdue payments or finding the right market segments. Teams risk running scattered projects without a coherent set of problems, objectives and priorities. This means:

• starting with clear business priorities,

• ensuring data hygiene is rock-solid,

• and remembering that the newest AI isn’t always necessary.

Small or midsize companies can absolutely benefit just by blending internal data with external sources and staying open to simple, problem-first approaches before diving into complex automation.

Ultimately, throwing AI at the wall to see what sticks is a losing battle for ROI. The real magic happens when you put data to work toward clear business goals.

Ready to Explore Data & AI for Your Organization?

At Agilytic, we thrive on practical problem-solving. Our data and AI consultancy services have helped companies of all sizes streamline processes, boost sales, and reduce operational risk. We’d love to learn about your goals and see if there’s a data-driven path to meet them.

• Schedule a Call: Let’s discuss your challenges, see where data can make a difference, and decide if a simple analytics project or a more advanced AI approach fits you best.

• Listen to the Full Podcast Episode: Dive deeper into Julien’s stories, available on most podcast platforms and YouTube.