Harnessing the potential of AI effectively requires discernment—and sometimes, a serious dose of self-reflection.
The EU’s Digital Decade Report tells us that one in four Belgian companies had adopted at least one AI application in 2024. But adoption is far from uniform: while two out of three large companies have embraced AI, only 23% of SMEs have done so. In recent months, Aurélie Couvreur, head of the digital adoption facilitator MIC, has observed a rapid increase in maturity. “Where many were still trying to understand what ChatGPT or Copilot could do, they now recognize that AI drives productivity and innovation,” she explains. More and more companies are exploring the setup of AI agents, particularly conversational robots (chatbots).
Finding Real Added Value
Administrative management, pitching, customer interactions, marketing campaigns, recruitment, training, finance… AI applications are gradually becoming commonplace across all company functions, making certain tasks easier. But that doesn’t automatically translate into a game-changing impact. “Too often, AI is used superficially to do ‘more of the same,’” notes Gaëlle Helsmoortel, Business & AI Strategist at Dataroots. Flooding social media with posts just because they are easier to write—over 50% of LinkedIn posts are now AI-generated, according to several studies—won’t necessarily boost sales if the content lacks originality. Similarly, AI embedded in software suites—like taking meeting notes and sending ‘to-do’ lists—saves time but provides only limited added value.
AI drives productivity and innovation.
- Aurélie Couvreur
“AI will be far more impactful when, after six months, it can process the mass of documents such as objectives, briefings, reports, and meeting transcripts. It can then help assess project progress, identify bottlenecks, and suggest solutions or next steps,” Couvreur continues.
She confirms: “Adopting AI purely for productivity isn’t the right approach, because sooner or later, all your competitors will do the same. The real competitive advantage lies in using it to boost innovation or service quality.”
Defining and Managing Your Project
How can companies successfully carry out such projects? Start by clearly identifying the operational challenges you want to address—long customer service wait times, recurring administrative errors, time-consuming email campaigns—and ensure AI can address them efficiently and cost-effectively, which often requires a critical mass of data.
Evaluating a project also requires reliable performance indicators to set clear AI objectives (e.g., “How many calls does my call center handle today, and can AI improve that number?” or “Can AI process a list of prospects effectively?”). This entails gathering accurate, sufficiently available, correctly labeled, and practical data.
Next, structure your process. AI projects involve multiple feedback loops based on data analysis, learning, and adjustments. Once your pilot project succeeds, ethical considerations must guide applying its lessons to other departments or integrating them into existing systems—ERP or CRM, for instance.
Budgeting as a Strategic Investment
This requires resources: paid versions of tools, time for developing proprietary solutions, recruiting internal or external data specialists, team training, change management, and AI model maintenance. It’s easy to hesitate if the return on investment is unclear. “That’s why AI should be considered a strategic investment. It’s not about imposing it everywhere, but rather identifying the areas where it can truly make a difference,” stresses Aurélie Couvreur.
Redefining Your Value Proposition
Translation, legal or management consulting, communication… In many sectors, AI has the potential to profoundly disrupt—or even replace—the core business. In a world where merely reproducing knowledge no longer constitutes value, reinvention becomes essential. But how? A translation firm, for instance, can leverage expertise built through long-term client relationships, particularly technical glossaries developed over time, often better mastered by external translators than the clients themselves. Consultants can repackage their knowledge into tailor-made training sessions.
Yet with AI producing a “first draft,” each translation or presentation slide requires far less work, and clients are aware of this. “A fee-based system becomes disadvantageous. If you claim not to use AI, clients may see you as outdated. If you confirm its use, they may demand a discount,” notes Gaëlle Helsmoortel. This is why moving to a value-based pricing model, for example through flat fees, becomes necessary.
Couvreur also cites a law firm specializing in traffic offenses. Using an internal database with years of expertise - rulings, fines, and pleadings - AI produces an initial response free of charge, guiding clients toward the most appropriate legal procedure or settlement. Clearly, embracing AI often means embracing deep self-reflection.
Beci supports the digital development of companies. Discover the projects of our Digital/AI community.