For about fifteen years, the word "digitalization" held together a variety of operational meanings that today it would be useful to separate. For some companies it meant computerizing paper processes. For others, getting an ERP. For yet others, opening a website, launching an e-commerce, activating social accounts. For others still, automating repetitive activities with software tools. All these things are indeed part of digitalization, and there are industry blogs from the last fifteen years that have told all of them.
What has changed in the recent period is that the frontier of what it means to "be digitalized" has shifted significantly, and much of what until recently was considered the digital vanguard is today assumed as the operational baseline. Having a website is no longer digitalization — it's a requirement of commercial existence. Having a CRM is no longer advanced digitalization — it's standard practice for anyone managing structured commercial relationships. Having process automations is no longer innovation — it's the minimum level to stay competitive on costs.
Contemporary digitalization is something different, and it's worth articulating it for what it is now, not for what it was a decade ago. It articulates on three levels that deserve separate treatment: a level of basic digital infrastructure, by now taken for granted in any functioning company; a level of structured integration and automation, which is the operational terrain where many companies are currently working; and a level of integrating generative artificial intelligence into business processes, which is the frontier where the new competitive divides are being drawn.
Understanding where your company sits relative to these three levels is probably more useful than reading yet another generic list of "the benefits of digitalization."
Level one: the basic digital infrastructure
The first level includes what most industry blogs continue to present as "digitalization." Integrated management systems (ERP), customer relationship management software (CRM), internal communication and collaboration tools, a structured web presence, any e-commerce platforms, digital document archiving systems.
It's worth naming it for what it is now: an operational level that the Italian companies competing on serious markets built long ago, and that no longer represents a competitive differentiator but a baseline of existence. A medium-sized company without a working ERP has today a significant operational problem. A B2B company without a structured CRM is working with lower efficiency than its competitors. A B2C company without a curated digital presence is leaving commercial volumes on the table that translate into lost revenue.
For Italian SMEs that still have gaps on this level — and they exist — the priority is to close the gap quickly, with mature tools, consolidated processes, reliable vendors. It isn't a strategically complex activity, it's an operationally demanding activity that requires planning and discipline. There exist in Italy ecosystems of consultants, system integrators, established management-software vendors that have made this level accessible even to the smallest companies. National and European public funding tools regularly make specific resources available for this phase.
The point not to confuse is this: completing level one doesn't mean "being digitalized." It means having arrived at the point from which true contemporary digitalization begins.
Level two: structured integration and automation
The second level is where many companies are working today, and it's the terrain to which the competitive differentiators are shifting.
Integration between different systems is the first theme. A typical medium-sized SME has the ERP, has the CRM, has a management system for logistics, possibly has a tool for customer service, has accounting systems, has the e-commerce presence. All these systems often work in parallel but poorly integrated. The data that should flow from one system to another is often transferred manually, with the risk of errors and with times that reduce overall operational speed. Structured integration — through APIs, middleware, integration platforms — is what makes it possible to transform a collection of separate systems into a unified infrastructure that works coherently.
Process automation is the second theme. Not the automation of single activities — many companies already have point automations — but the automation of workflows that cut across multiple business functions. A customer order that automatically converts into production planning, inventory management, logistics scheduling, invoicing, accounting entry, commercial follow-up. These are levels of automation that require structured process design and investment in specific tools (RPA — Robotic Process Automation, workflow automation platforms, business process management systems).
The analysis of structured data is the third theme. The companies that have completed level one have accumulated significant quantities of operational data over the years: sales history, customer behaviors, purchasing patterns, production performance, logistics data. Transforming this data into information usable for strategic and operational decisions is a level of work that requires specific tools (business intelligence systems, dashboarding, data analytics), dedicated internal skills, and a corporate culture that actually uses data in decisions.
Structured cybersecurity is the fourth theme, increasingly central. With the increase in operational dependence on digital systems, the protection of these systems has become a critical function. For SMEs, this level is often undersized, with protection structures that don't correspond to the level of risk they're exposed to. The European NIS2 regulation has significantly extended cybersecurity obligations to a broader perimeter of companies, making the theme no longer postponable for those who fall within its scope of application.
For the companies that have completed level one, the work on this second level is where the most relevant operational investments of the current moment are found. They aren't spectacular investments — they don't produce newspaper headlines, they aren't visible on the surface — but they produce cumulative efficiencies that translate into margins, operational agility, capacity for scale. They're less glamorous investments than the third level, but equally strategic in the medium term.
Level three: integrating generative artificial intelligence
The third level is the contemporary frontier. The arrival of generative artificial intelligence technologies — language models like Claude, ChatGPT, Gemini, and the systems for generating images, audio, video, code — has opened operational possibilities that until recently didn't exist.
Unlike levels one and two, where the technology is by now mature and implementation practices are consolidated, level three is still in a phase of active exploration. The companies integrating it seriously are experimenting, discovering use cases, building skills, modifying processes. It's a phase that requires operational curiosity, willingness to invest in learning, acceptance that some experiments won't produce the expected value.
Four areas are currently the ones in which generative AI integration produces tangible operational returns for companies.
Content production. Generating commercial texts, marketing materials, technical documentation, product sheets, translations, communications for customers is an activity where AI tools significantly reduce times and costs while requiring human supervision and review. For companies that produce significant volumes of content, integrating AI pipelines into editorial production is one of the use cases with the fastest return.
Analysis and research. The analysis of complex documentation — contracts, industry reports, regulations, communications with customers — is an area where AI tools make it possible to handle volumes of information that human work alone couldn't process in the required times. Preliminary market research, competitive analysis, regulatory verification are activities that can be significantly accelerated.
Customer service and the customer relationship. AI conversational systems integrated into the customer communication channels can handle a significant share of requests autonomously, leaving human operators only the complex cases or those that require judgment. The quality of the interactions is today sufficient to be actually useful in many contexts, especially if the AI is well integrated with the company's information systems (CRM, management systems) to access customer-specific information.
Software development and technical management. For companies that have significant software development components (even the SMEs that manage their own portals, applications, integrations), AI programming-assistance tools have significantly increased developer productivity. The ability to prototype, test, develop new functionalities is progressively accessible to smaller teams than were needed until recently.
Beyond these four areas, AI is progressively entering more specific functions — human resources management, financial analysis, supply chain, predictive maintenance, quality control — with applications that vary by sector and by technological maturity.
The important thing to understand about level three is that it requires a different investment from that of the previous levels. It isn't enough to buy tools — you have to integrate the tools into the processes, train the people, build internal skills on prompt engineering and on the strategic use of AI technologies, reconfigure the workflows to exploit the new possibilities. It's work that requires time, experimentation, acceptance of the fact that the first applications might not be the definitive ones.
The relationship between digitalization and company size
A question worth addressing is the relationship between the level of digitalization and the size of the company. For years industry blogs have told digitalization as a theme mainly for large companies, with generic references to the "necessity for SMEs too." The contemporary reality is more nuanced.
For level one (basic infrastructure), the tools are today accessible to companies of every size, with proportionate costs that allow even micro-enterprises to equip themselves with management systems, CRM, curated digital presences. The gap between large companies and SMEs on this level is today more a matter of strategic choice than of technological accessibility.
For level two (integration and automation), the gap between large companies and SMEs exists and tends to be significant. Large companies have dedicated internal teams, budgets for sophisticated systems, the capacity to integrate complex tools. SMEs generally work with more contained resources and with multiple priorities. The accessible tools have grown — many automation and integration platforms have versions accessible to small companies — but effective implementation requires internal skills or reliable external consulting.
For level three (generative AI), the picture is paradoxically different. Generative AI technologies have made accessible to small companies levels of operational capacity that until recently were reserved for the large players. An SME with a team of six people can today produce content, do analysis, manage communications with customers at levels that twenty years ago would have required a team of thirty people. It's one of the few cases in which the technological frontier is more democratizing than concentrating — at least in the current phase.
This means that Italian SMEs today have a specific opportunity to close scale gaps with larger competitors through the intelligent integration of AI technologies. It's an opportunity that requires, however, the will to experiment, invest in internal skills, modify consolidated processes. It doesn't materialize by buying software licenses — it materializes with a transformation path that has its own timing.
Incentives and support tools
For Italian companies planning investments in digitalization, there exist public support tools worth knowing. They've changed over time and will change again, but the structural principle is that investment in digital transformation has been supported by Italian and European public policies for several years, and this is destined to continue.
National investment plans dedicated to digital transformation, fiscal measures for investments in advanced technologies, vouchers for innovation consulting, European funds dedicated to the digitalization of SMEs — these are families of tools that have succeeded one another over time under different names and that will probably continue to evolve. The specific picture of the moment should always be verified with up-to-date consultants, because it changes frequently.
The practical point is that planning a significant investment in digitalization is worth structuring also from the point of view of the available incentives. Specific consulting on these themes is generally accessible through trade associations, chambers of commerce, specialized consultants. For SMEs, financial planning that integrates incentives and their own investments can significantly reduce the net cost of the transformation.
The human dimension, which remains central
It's worth dedicating attention to a dimension that narratives about digitalization tend to neglect: the human factor.
Digitalization isn't mainly a technological activity — it's an activity of organizational transformation supported by technologies. The most sophisticated systems produce limited returns if the people who are supposed to use them aren't trained, aren't motivated, aren't integrated into the vision of the change. The companies that have realized successful digital transformations have invested significantly in training, in change management, in building internal skills.
For Italian SMEs, this means planning technological investments together with investments in training and accompanying people. It means assessing the capacity to absorb change before launching overly ambitious projects. It means involving people in the transformation processes, listening to their observations, modifying the plans according to what emerges in practice.
A recurring difficulty in the digitalization of SMEs is the underestimation of the human work needed to integrate technology into real processes. Companies that have bought sophisticated systems that remain underused because the people haven't integrated them into their daily work are a frequent situation, and they tell something important about the nature of digital transformation.
What makes digitalization actually useful
To close on a practical note, it's worth articulating some characteristics that distinguish digitalizations that produce value from digitalizations that produce mainly costs.
They start from real problems, not from solutions looking for a problem. The digitalizations that work are born from a concrete operational difficulty that technology can help solve. The ones that produce less value are born from the feeling of "having to digitalize something" without a specific direction.
They're planned in phases. The complete digital transformation of a company is a multi-year path. Trying to do everything at the same time is a guarantee of confusion and fragmentary results. Choosing one or two priorities, completing them, moving on to the next — that's the operating pattern that works.
They measure the results. A digitalization that isn't measured is a digitalization that probably isn't producing the expected returns without anyone noticing. Defining beforehand what "success" will mean for a specific investment, and then verifying it with concrete data, is operating practice that many companies neglect.
They invest in people alongside the technology. As said above: without the human factor, technology produces partial value. The companies that recognize this structurally obtain more consistent returns.
They maintain flexibility. Digital technologies evolve rapidly. Tools that today seem definitive can be superseded in a few years. Structuring your digitalization so that the components are replaceable and updatable — rather than tying yourself to closed proprietary solutions — is a form of strategic prudence that pays off in the medium term.
Digitalization isn't one single thing — it's a path articulated on different levels, with priorities that change according to the company's starting point. The Italian companies that plan technological investments with awareness of their current level and the level they want to reach build transformations that produce value.
The practical thing to do, for an Italian company that wants to reason seriously about its digitalization, is probably this: honestly assess where it stands today on the three levels described, identify the most relevant gaps, plan specific interventions with realistic timing, integrate technological investments and investments in people, maintain flexibility for the adjustments that will become necessary along the way.
Digitalization produces value when it's a tool at the service of a clear company vision. When it's an end in itself, it tends to produce costs without the proportionate returns that had been imagined.