Has your company already settled into the seats of first class, or have you not yet found your place on the AI train?
Our AI expert Alexander Frimout explains which processes are ideally suited for your first AI business case.
And why some companies aren’t sitting in the best wagons…
The possibilities are enormous. The use cases that prove this are expanding at a rapid pace. Many companies therefore feel the pressure to do something with AI.
But what exactly is that something? Where do you start as a company with your first AI business case? How do you identify the tools that truly add value to your business? Or put differently: in which wagon of the train below should you take a seat?
For many companies, that means the doors of the first wagon right behind the locomotive: chatbots.
Why there?
Probably because the ChatGPTs and Geminis of this world are the AI technologies we feel most familiar with. Of all the tools, platforms, and apps, chatbots are the most widely known to the general public. That makes the threshold to use them in a professional context just a little bit lower.
Yet this very wagon is the least interesting for companies, according to Alexander.
“Chatbots improve your personal efficiency. But their impact at company level? Limited. I’ve never spoken with a company that could present a concrete return on investment.”
“And yet, this is exactly what happens in many companies: we roll out 100 licenses of Microsoft Copilot and call it a day. What those colleagues actually do with it remains vague. As a company, you can tick the AI box, but in my view, it doesn’t deliver real value. It’s not a полноценный business case with a clearly defined ROI.”
According to Alexander, the real gains lie at an intersection. The intersection between the last four wagons and back-office applications.
First, a quick explanation of those wagons.
Technology: check. Now let’s look at the business processes that are most relevant for your very first AI business case.
The most interesting AI business cases are concentrated in back-office applications that are well integrated with existing software.
Think of LLMs that help your customer service teams respond faster. Or custom software that scans delivery notes and automatically converts them into a standard template.
Why are these processes the most interesting?
Let’s make that intersection between AI technology and back-office processes more concrete.
Meet Decomecc. This company from Genk processes aluminum and (stainless) steel: cutting to size, sawing, applying coatings…
At their site, trucks constantly come and go, delivering coils like the one in the image above. Decomecc processes around 20 to 30 deliveries per day.
The situation before…
Administrative staff had to check every delivery and delivery note:
This meant: reading the delivery note, searching in the database, manually retyping the information…
In the best case, this takes about 15 minutes. But not every driver immediately has the correct paperwork. And not every colleague finds the right data just as quickly. On top of that, manual work always comes with a margin of error.
The situation now
The employee scans the delivery note. It is automatically sent to the custom AI tool we built for them. The tool then searches for the correct information and matches it with the internal database.
Within a minute, the employee receives a proposal:
As a result, the employee now has time to focus on the delivery notes the AI tool cannot immediately identify. For example, because the delivery note is incorrect, the driver delivered on the wrong day, or the order itself is wrong.
The end result?
🎥 Watch the full video to see how we helped Decomecc automate their administrative paperwork.
Do you already have an idea which back-office processes would be ideal for your first AI business case? The more boxes you can tick on the checklist below, the more relevant it becomes.
All boxes checked? Then we’ll help you in the next step to determine the ROI and translate your case into a technical solution.