Latest / Scaling Up with Fexingo: How Small Businesses Become Mid-Market Companies / How a Niche Audio Brand Scaled Past 15 Million by Fixing Demand Forecasting
Transcript
- Lucas: So there's a moment in every growing product business where the founder realises the problem isn't sales. It's everything behind the sales. Today I want to talk about a boutique headphone maker that hit 15 million in revenue and then got stuck — not because demand fell, but because they couldn't predict it. Luna: I love that. Because everyone thinks scaling is about getting more orders. But if you can't forecast, more orders just means more chaos. Lucas: Exactly. And this founder — former audio engineer, started in his garage — he thought his problem was marketing. He was spending more on ads, but his warehouse was full of the wrong stuff. Turns out his forecasting was broken. And that's what we're digging into. Luna: Before we get deep into the numbers — we should say, we don't run ads on these episodes. That's a choice. If these conversations have helped you think about your own business, and you want to support that, the link is buy me a coffee dot com slash fexingo. No pressure, just keeps it independent. Lucas: Yeah, that's a nice way to put it. And it's genuinely appreciated. But back to the headphones — the story is a masterclass in why demand forecasting can make or break a scale-up. Lucas: So this company — let's call them Acoustic Labs, not their real name — they had a flagship wireless headphone that was selling well. But they also had a dozen other SKUs that were eating cash. The founder told me he could feel the business getting heavier, but he couldn't figure out why. Luna: Heavier. I like that. So what did he do first? Lucas: He did what most founders do — he doubled down on sales. More influencer deals, more discounting. But the margin pressure got worse. So he finally pulled the inventory report and saw the truth: his forecast error on some SKUs was over 60 percent. He was ordering based on gut feel, not data. Luna: Sixty percent error — that's not a forecast, that's a guess. So what was the fix? A software tool? Lucas: That's the interesting part. He tried a bunch of expensive forecasting software, but the real fix was simpler. He set up a weekly forecast review. Every Thursday, he sat down with his sales lead and operations person, looked at the last four weeks of actuals, and adjusted the next eight weeks of orders. It took twenty minutes. Luna: Twenty minutes a week. And that was enough to turn it around? Lucas: It was, because the discipline forced them to notice patterns. They realized their flagship headphone had a huge spike in demand every time a certain tech reviewer posted a video. And they started building that into their forecast. Within a quarter, their forecast error dropped to under 25 percent. Luna: Still high, but much better. So what did that do to the cash flow? Lucas: Huge. Their inventory write-offs went from about 8 percent of revenue down to under 3. That alone freed up over 700,000 dollars a year. And with better cash flow, they could finally ask their suppliers for net-60 terms instead of being stuck on net-30. That gave them another working capital boost. Luna: So the forecast wasn't just about avoiding overstock. It unlocked better terms from suppliers because they were a more predictable buyer. Lucas: Right. And that's the thing — a supplier will give you better terms if they trust that you'll order consistently. This company went from 'we'll ship when we can' to 'we can build a schedule around you.' That's a massive shift in negotiating power. Lucas: But the founder also told me about the one time the forecast failed spectacularly. They launched a limited-edition colorway, and it sold out in three days. They thought it would take a month. So they scrambled to reorder, but the lead time was eight weeks, and by the time the stock arrived, the hype had died. Luna: Ouch. So they missed the wave. What did they learn from that? Lucas: They started doing pre-orders for limited drops. That way they only manufactured what was already sold. It's a classic strategy, but it only works if you have a forecasting cadence that can support it. They did, so it worked. Luna: I think a lot of listeners might be thinking, 'We're not a product business. Does this apply?' But I'd argue every business has a form of inventory — even a service business has capacity. Lucas: Totally. For service businesses, it's about forecasting your team's capacity versus incoming demand. Same principle. Under-forecast and you've got idle staff. Over-forecast and you're scrambling to hire at the worst time. Lucas: So the takeaway from Acoustic Labs is not that they invented some new algorithm. It's that they built a simple, repeatable process and stuck to it. And that process gave them the clarity to make better decisions across the whole business. Luna: It's almost boring how simple it is. But the results are anything but boring. I mean, going from 15 million to 22 million in two years — that's real. Lucas: And the founder's now the first to admit that the problem was never sales. It was the fear that if he sold too much, he couldn't deliver. Now he can plan for growth instead of just reacting to it. Luna: So for our listeners who want to try this, what's the first step? Just start a spreadsheet? Lucas: Start with your top five SKUs, or your top five services. Pull twelve months of history, find the seasonality, and then build a simple moving average forecast. Don't overthink it. The goal is to start the habit of comparing forecast to actual every single week. Luna: And if the forecast is off, you adjust. Over time, you get better at spotting the signals that predict demand. Lucas: Exactly. And once you've got that discipline, you can start asking the bigger questions — like, should we take on that big order that might blow up our inventory? With a forecast in place, you can actually answer that without crossing your fingers. Luna: That's a good place to leave it. Thanks for listening, everyone. Lucas: Yeah, and if you want to support the show, the link is buy me a coffee dot com slash fexingo. See you next time.