The word "digitalisation" in Italy has come to a bad end. It was appropriated first by software suppliers, then by process consultants, then by marketing agencies, then by regional grant schemes handing out digitalisation vouchers. Each step emptied it a little. Today it means so many things that it no longer means anything.
This article does not talk about public grants. It does not talk about regional vouchers. It does not talk about digital transformation as a slogan. It talks about something concrete many SME owners observe after years of investment: they already have, inside the company, most of the tools. Yet something does not work as it should. And it is not another tool that solves it.
The real state of Italian SMEs
An Italian SME of medium size, say between ten and fifty employees, typically has an administrative management system (one of the known names on the market). It has a company email inbox for each collaborator with its own domain. It almost always has a website. It often has a CRM, even if not always updated. It has a LinkedIn page and probably a Facebook or Instagram page. It has one or more WhatsApp groups for internal communication. It has access to Google Drive, Dropbox or OneDrive for file sharing. It has, for a year or two now, some subscription to AI tools (more often ChatGPT Plus than the alternatives).
This set-up, from a purely technical point of view, is much richer than what the same type of SME had ten years ago. In theory, everything is in place. In theory, one should be able to work much better than one used to.
In practice, in the same SME, the owner is still the last to know the important things. Internal communication is chaos distributed across WhatsApp, email, oral meetings, phone calls. The younger collaborators look for information in the wrong places because they do not know where it is. The seniors answer the same questions ten times a month. Important decisions wait because those who can make them are overwhelmed. Clients receive experiences that vary depending on who replies that day. The recent AI tools are used by one or two individuals who get something out of them, while the rest of the company continues as before.
The gap between the theoretical set-up and real functioning is enormous. And it cannot be bridged by buying another tool. On the contrary: those who try to bridge it with another tool usually make the problem worse, because they add one more piece to a system that was not in order.
The problem, looked at squarely, is that the tools do not work together because what should hold them together is missing. Not a technical integration, which is often present and well done: an informational and cognitive integration.
A concrete example. A client emails a salesperson asking to rework a proposal. The salesperson replies by email with the updated proposal. In the internal WhatsApp group, he tells colleagues "I've redone the proposal for Rossi". In the CRM, nobody updates anything because the salesperson thinks someone else will, and the other thinks the opposite. The proposal is accepted. The client asks for a contract. The administrative manager has seen neither the email nor the WhatsApp message and asks the salesperson to send him the details. The salesperson, tired of repeating, forwards the email. The administrative manager prepares the contract based on the old proposal (the one archived in the CRM, not the new one in the email). The contract arrives wrong. The client complains. The salesperson writes an apology email.
This sequence, trivial in its components, is the daily fabric of many SMEs. Every single tool did its job. The email delivered the message. WhatsApp relayed internal communication. The CRM archived. The management system produced the contract. And the result was an error that cost time, image, and part of credibility.
The problem is not that a software is missing. The problem is that structured information flows tying together email, CRM, management system and internal communication are missing. These flows are work of organisation, not of technology.
What structuring the flows means
Structuring an information flow means answering concrete questions about the company's typical day. It is not a theoretical exercise. It is an accurate observation of what already happens, compared with what should happen, with the delta between the two becoming the basis of the work.
When a client sends a request by email, what happens? Where is it recorded? Who is notified? How do we understand if it is a new request or the continuation of something existing? In the example above, the correct flow would have been: email with the rework request, the system associates it with client Rossi in the CRM, the salesperson records the rework directly in the CRM instead of in a new email, the administrative manager, before preparing the contract, consults the latest version in the CRM and not an email.
If this flow is not defined explicitly, each collaborator follows his own preferences. Someone records in the CRM, someone does not. Someone notifies on WhatsApp, someone by email, someone does not notify. The company as a whole behaves inconsistently, and the client receives experiences depending on who serves him at that moment.
Structuring the flow does not mean imposing a rigid procedure. It means defining the basic rules of how information circulates, so that every collaborator knows what to do with a piece of information and where to record it. The aim is not control; it is that the system (people plus tools) works as a system, not as a set of individuals collaborating by consensus.
Mapping before buying
Before introducing another tool, a sensible company does one thing: it looks at its current information flows. It maps how information arrives, where it is recorded, who receives it, how it is used. It identifies the break points: where information is lost, where it is duplicated, where it reaches the wrong actor or arrives late.
From this map two types of problem emerge. Some are organisational: the flow is unclear, nobody has defined the rules, each person does it their own way. They are solved with an explicit definition and internal training, without technology. Others are structural: the flow is clear, but one of the tools is weak or does not talk to the others. They are solved with specific technological choices.
In most Italian SMEs, seventy to eighty per cent of "digitalisation" problems are of the first type. Tools are not missing; rules are. They are solved with a month or two of structuring work, without any new purchase. The remaining twenty to thirty per cent actually requires technology investment, but these investments are much more targeted and effective because they have a clear map of what is needed in front of them.
The opposite (buy first, structure later) is the pattern that produces the thirty thousand euros spent without return, which many owners have experienced.
The trap of language
Part of the problem is linguistic. The term "digitalisation" suggests a technological problem. Consequently, the expected solution is technological: another software, another platform, another integration. This expectation is cultivated on the supply side: suppliers and technical consultants have an interest in presenting SME problems as solvable with their tools.
An owner who reformulates the problem as "how information circulates in the company" shifts to a different plane. It is not a tool problem; it is an organisation and knowledge problem. It is solved in a different way, with different actors, and with different costs. Often lower than what would have been spent on the next tool.
This reformulation is not marketing. It is the key to avoiding the trap of recurring spending on tools that do not make the difference. After a few cycles of technology investment without results, many owners begin to suspect that the problem is not technological, but they do not know how to formulate it differently. The language proposed to them from outside does not help.
An exercise that takes less than an hour
If you want to understand where you are in your company, an exercise lasting less than an hour produces a clear picture. Take a typical transaction (a new client, a recurring request, a service incident) and go through step by step what happens. Where does the initial information arrive? Who receives it? Where is it recorded? Who else receives it as a consequence? Where does it end up? What does each person do with it?
If, while telling the journey, you notice you often use the words "usually someone", "it depends on who", "normally we pass the message on", you are looking at an unstructured flow. There is no explicit rule; there is an uneven habit. This is the ground on which digitalisation done starting from tools fails.
The work that counts is making these flows explicit, understanding which are in order and which are not, deciding what to structure and what to leave flexible. Only afterwards do you return to tools, with much more targeted choices.
It is a type of diagnosis better done with someone looking from the outside, with method, without assumptions about which tool is the solution. In forty-five minutes of structured conversation, a real operational map emerges, which stays with you and guides any future choice, even if you decide to deal with it at another time.