Table of contents
Introduction
The return on investment (ROI) of artificial intelligence has become a pressing issue for businesses following an article published by TechCrunch on 25 October 2026, which reports on an analysis by Torsten Slok, chief economist at Apollo Global Management: outside the major tech companies, AI has not yet had a visible financial impact on margins and revenue per employee. For Italian entrepreneurs, CEOs and marketing directors, this means one simple thing: from 2026 onwards, it is no longer enough to say ‘we’re using AI’; you need to understand where it actually fits into the business.
The issue is not whether artificial intelligence is useful. It is, and in many areas it is already changing the way content, campaigns, analytics and decision-making processes are designed. The point is that it is one thing to use a tool, but quite another to turn it into value for the business.
We often see this when talking to companies in the fashion, food, beverage, home and furnishings, and manufacturing sectors. There is enthusiasm, there is experimentation, and there is software that promises efficiency. However, when you look at the profit and loss account or the quality of sales leads, the return on investment is not always immediate.
That is why, at Dandicom, we treat AI as a component of our strategy, not as a shortcut. Our AI solutions for marketing only make sense when they work in tandem with positioning, creativity, advertising, content, data and measurement. On its own, AI produces outputs. Within a project, it can help to create value.
Data from TechCrunch and Apollo on the resurgence of AI
The TechCrunch article quotes a note from Apollo Global Management in which Torsten Slok observes that, outside the major tech companies, there is still no evidence of an improvement in profit margins and revenue per worker attributable to artificial intelligence. This refers to the so-called S&P 493, i.e. the S&P 500 excluding Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla.
This article is interesting because it shifts the focus from stories about innovation to business figures. It does not discuss how many people use AI tools or how many companies have tested them, but rather productivity, revenue and profit margins. This is an important distinction, particularly for those running a business with specific commercial objectives.
According to Slok, AI may be a powerful computing tool, but in business, intelligence is above all about judgement. Trust, corporate culture, accountability, unwritten rules, reputation and ethical decisions have just as much of an impact on results as the ability to process information. In practice, software can speed up certain tasks, but it cannot replace strategic leadership.
This is precisely why data only becomes useful when it is interpreted. A company may have very comprehensive reports and still fail to make better decisions. This is why AI-driven data analysis and insights are effective when they help management to identify signals, priorities and scenarios, not when they simply add yet another dashboard to monitor.
The ROI of artificial intelligence: beyond the novelty effect
The first mistake is to rely on the novelty factor to drive the success of AI. It may work for a few months: content generated in a new format, a striking image, a campaign that conveys innovation. But the market’s attention quickly wanes. What remains is the quality of the project.
In corporate marketing, the ROI of artificial intelligence does not arise simply because a tool writes text faster or generates visuals more quickly. It arises when that output is consistent with the brand’s positioning, speaks to the right audience and forms part of a measurable communication strategy. Without this link, AI risks producing content in large quantities, but not necessarily useful content.
Consider a company in the furniture sector that wants to publicise the launch of a new range. AI can help explore creative options, visualise settings and speed up certain stages of production. But without a clear creative direction, the result may look good but fail to distinguish the brand.
That is why we talk about AI for marketing and communication as an integrated process. It requires communication skills, knowledge of the channels, an understanding of the appropriate tone of voice, and the ability to interpret the market. Artificial intelligence expands the possibilities, but it does not decide on its own which message deserves to reach the customer.
Where AI can create value in corporate marketing
There are areas where AI can have a tangible impact on marketing work, particularly in companies that already operate in a complex commercial environment. It is not magic, but rather a tool to support processes that often require time, consultation and numerous variations. Here, artificial intelligence can act as a catalyst, provided it is guided by clear objectives.
In content marketing, for example, it can help to organise ideas, tailor messages to different audiences, and make the transition between strategic and operational content smoother. In a food or beverage company, this might mean ensuring consistency between brand storytelling, seasonality, marketing campaigns and sales materials. Consistency, however, remains a human choice.
In advertising, AI can help interpret signals, generate creative variations and manage more dynamic contexts. But a campaign does not thrive on automation alone: it thrives on positioning, the offer, creativity, the landing page, data quality and the ability to interpret what happens after the initial contact. AI initiatives for advertising campaigns work when they remain connected to the company’s entire digital ecosystem.
Then there is the issue of visual production. AI-generated or AI-assisted photos and videos can open up interesting possibilities, particularly for concepts, visualisations and digital content. But for a fashion, home or industrial brand, the credibility of the image still matters immensely. Content that is too generic, even if technically accurate, can weaken the perception of the brand rather than strengthen it.
Why AI investments do not yield immediate returns
Many investments in AI do not pay off straight away because they are treated as a technology purchase rather than a business project. A tool is chosen, introduced into a department, and an improvement is expected. But that improvement rarely materialises unless it is first made clear which marketing, communication or sales problem one is seeking to address.
Another reason is the gap between experimentation and measurement. A team may use AI to produce more content, but if it is not clear which content boosts brand awareness, which helps generate leads and which drives sales, the return on investment remains unclear. You work harder, but you don’t always work better.
There is also the issue of internal governance. In some companies, AI is used by different people with different approaches, without a shared vision. The result is inconsistent output: texts with varying tones, images that lack coherence, and data interpreted in a one-sided manner. For a well-structured brand, this fragmentation can become a hidden cost.
It’s a problem similar to the one we see when a company invests in digital without a plan. Channels are active, campaigns are running, content is being published, but there’s little connection between the different elements. We’ve also discussed this in relation to investing in digital without a strategy: technology amplifies whatever it finds. If it finds confusion, it often amplifies that too.
What does this mean for Italian companies?
For Italian companies with an already well-established structure, AI should not be seen as a race. It should be a choice driven by digital maturity. There is no need to adopt it simply because everyone is talking about it; rather, it is important to understand how it can improve the way the brand communicates, attracts attention, interprets data and supports commercial activities.
This is even more true for businesses with a strong product identity. The Italian fashion, furniture, food, beverage and manufacturing sectors do not simply sell technical specifications: they sell reputation, taste, reliability, relationships and perceived quality. These are elements that AI can help to convey, but which it cannot invent from scratch.
In our 22 years of working in marketing and digital communications, we have seen many innovations only become truly useful once they have been incorporated into long-term partnerships. It is there that increased brand visibility, higher-quality content and greater consistency between communications and commercial objectives are built. AI follows the same logic: it is not a one-off, but a tool to be integrated.
The ROI of artificial intelligence, therefore, should not be sought in the promise of an immediate return, but in its ability to link technology, strategy, creativity and measurement. If you want to understand where AI might be useful for your business, we can discuss this by starting with real objectives rather than the tool of the moment, through our AI consultancy service for businesses.
FAQ
ROI compares the return generated by an investment with the cost incurred in making it. In the case of AI, however, the assessment should not be limited to time savings: data quality, the impact on processes, communication and the contribution to business objectives are also important.
For a business, using artificial intelligence means first understanding where it can create real value: marketing, content, data, advertising, automation or decision support. The choice of tools comes later, because without a clear direction, AI risks remaining a one-off experiment.
ROI measures the return on an investment relative to the capital employed. ROE, on the other hand, measures the profitability of a company’s equity. These are different indicators: the former looks at the quality of an investment, whilst the latter looks at the company’s ability to generate a return on its own capital.
It is likely that AI will become more integrated into business processes, be seen less as a separate tool, and play a greater role in day-to-day activities. For businesses, the ability to manage it effectively – using expertise, reliable data and a clear understanding of their market – will make all the difference.