How Ardo nailed its forecasts
Crunch makes companies more effective. For Ardo, this meant becoming even better at getting customers everything they need, when they need it. The industry lingo being 'On time, In Full' - or OTIF for short.
Imagine planning a big family dinner, but the supermarket only has 90% of your ingredients. No potatoes. Annoying, right? Now scale that problem to an industrial level. That’s the headache Ardo’s clients face.
To fix it, you have two options.
- Throw cash at the problem. Buy massive amounts of inventory and hope for the best.
- Get scarily good at forecasting demand.
Guess which one Ardo picked.
So, what makes a forecast useful?
Let’s be honest: a basic forecast is easy. A useful one is a different beast. Here’s how we build it.
Level 1: The ‘good enough’ forecast. (Spoiler: it isn’t).
Most tools just look at recent averages. It’s simple, quick, and never outrageously wrong. It’s also not particularly right. One look at the graph tells you everything you need to know: you’re leaving money on the table.
Level 2: The forecast that gets your business.
The next step? Better models. Any data science company can sell you those. The trick is building models that understand your reality.
For Ardo, that meant digging into their product lines, customer concentrations, and the nitty-gritty of their growing process and logistics. Without that context, you get models that look great in a slideshow but fall apart in the warehouse.
Get it right, and a realistic pattern emerges. Progress. At least on paper.
Level 3: The forecast that prepares for reality.
Being right on average is not the same as being prepared. When was the last time you had a perfectly average year? Exactly.
This is where we stop just predicting and start managing risk, using confidence intervals to map out what could happen, not just what we think will happen.
Level 4: Adding logic that makes the right call.
Here’s the thing: a better forecast doesn't automatically lead to better decisions. It can just be a more expensive way of getting things wrong.
Think your 95% service level is solid? Think again.
Here’s where gut feeling gets expensive. A customer orders five products. Your odds of getting that entire order perfect just plummeted from a seemingly safe 95% to a mediocre 77%.
That’s the razor-thin margin between a happy customer and a support ticket.
The right decision isn't about individual product forecasts. It’s about the whole order. It means using smart risk calculations to set stock levels that keep both your customers and your CFO happy.
The math gets hairy here. Which, for us, is where the fun starts. But we won't bore you with that right here.
The result?
A forecast built with the end goal in mind. One that turns data into decisions, and decisions into profit.
We don’t kiss and tell, so you won’t see exact numbers here. But we can say this: at Ardo, more customers are getting what they want, when they want it. And that’s a result you can take to the bank.
