Generational change: what to digitalize when the next generation takes over

Generational change: what to digitalize when the next generation takes over

Generational change digitalization in an industrial SMB means modernizing one process at a time, starting from the one that costs the most hours and gives the least visibility, not the newest or coolest one. The first process to digitalize is almost always the high repetitive volume one that today lives outside the ERP, in Excel, paper or email (data extraction from orders and invoices, reconciliations, recurring reports). The core legacy ERP (TeamSystem, Zucchetti, Odoo) should not be replaced because it is the accounting core that already works: the problem is what orbits it, not the system itself.

In brief:

• When a new generation enters an industrial SMB, digitalization should start from the process that costs the most hours and gives the least visibility, not the newest or coolest one.
• The legacy ERP (TeamSystem, Zucchetti, Odoo) almost never needs scrapping: the problem is the parallel Excel and paper orbiting it, not the accounting core that already works.
• The new generation expects zero paper, no double data entry and modern UX, while the old guard owns the exceptions no manual documents.
• You digitalize one process at a time, in parallel with the old one, without switching off production: at Soraia the first delivery is 4 weeks.
• The real risk of generational change is not technical but change management: bypassing the people who know the real processes instead of extracting their knowledge.

In the Biella textile district I’ve watched the same scene play out dozens of times. The son (or daughter) comes back after university or a stint abroad, opens the first drawer and finds a stack of hand-linked Excel files, paper orders, an ERP that only talks to whoever has run it for twenty years.

The temptation is to rebuild everything. It’s almost always the mistake that halts production and turns the two generations against each other.

What the new generation actually expects

It isn’t “AI”. It’s something more mundane and more reasonable: that work stops looking like 2005.

No printed sheet traveling desk to desk. No data entered three times into three different spreadsheets. Real-time visibility on orders, stock, margins without having to ask the one person who “has the right file”. An interface you can use from your phone, not just the shop-floor terminal.

This expectation is healthy. But if you turn it into “let’s rebuild the ERP from scratch”, you’ve just signed an 18-month project that will stop production and leave the old guard feeling dispossessed.

What NOT to modernize first

I’ll say it to your face, because it’s the mistake I see most.

• Don’t touch the accounting core that works. If the ERP closes the books and handles e-invoicing without trouble, that’s not where your problem is. Replacing it is the most expensive and least urgent project.
• Don’t digitalize the “coolest” process. The dashboard that looks great in a meeting recovers no hours. The Excel three people update by hand every day does.
• Don’t start without a baseline. Without knowing what a process costs today in hours and errors, you can’t know whether digitalizing makes sense, nor prove it to the skeptics.

What to digitalize first, in practice

The best candidate has three traits: high repetitive volume, output that must land in your systems, and today it lives outside the ERP (Excel, paper, email).

In manufacturing that’s often: data extraction from supplier orders and invoices, reconciliations, commission calculation, recurring reports for management. These are processes where an AI agent or a custom tool removes double entry and frees real hours, without asking anyone to change the ERP they use.

Then there’s the commercial side, often the most neglected. At ILTEC, an office technology firm in Biella, we rebuilt the AI-search-optimized website in 3 weeks and placed a commercial agent reachable via QR code right on the equipment: the customer scans, asks a question, the agent answers. It’s the kind of intervention the new generation grasps immediately and the old guard accepts because it touches nothing that already works.

How to manage two generations without a clash

The project rarely fails for technical reasons. It fails because the new generation arrives with the tool and bypasses the people who know the real processes.

The old guard carries the exceptions: the client who always pays at 90 days even though the contract says 60, the supplier with the odd line item, the job that skips a step. That knowledge isn’t in the ERP and isn’t documented. If you digitalize without extracting it, you build a system that fails exactly where it matters.

The sequence that works: first extract the real process from the people who run it, then digitalize it, then run it in parallel with the old one until you trust it. That’s precisely the point of an AI Adoption program: not just software, but the team that understands and owns it. If instead the need is an internal tool or a custom web app, we tackle it with custom software development, with the client owning the code from day one and no lock-in.

And if you want to understand how to train both generations on the same tool, start from how to structure corporate AI training.

The right pace

One process at a time. First delivery in 4 weeks, 30 days of hypercare, measurement against the baseline defined at the start. Then the next.

It’s not the fastest way to digitalize a company. It’s the way that doesn’t stop production and doesn’t drive away the people who’ve kept it running for twenty years.

CTA

If you’re joining the company and don’t know where to start, let’s talk: 20 minutes, no pitch. Or take the check-up in 3 minutes to figure out which process is worth digitalizing first.