Since November 2022, when ChatGPT was launched, Artificial Intelligence has entered the public debate and the everyday lives of students and professionals largely through chatbots. This has created the impression that AI is a recent technology mainly associated with generating text and conversations. In reality, AI has been part of our digital lives for much longer, often in less visible but highly practical forms.
For years, AI has powered the recommendation engines used by e-commerce websites to suggest products based on browsing and purchasing behaviour. It also enables social media platforms to personalise the content shown to individual users. Similarly, Artificial Intelligence lies at the heart of the “smart” technologies offered by digital advertising platforms, such as Google Performance Max and Meta Advantage+, which analyse vast amounts of data to optimise ad delivery in real time based on target audiences and budget allocation.
Chatbots are therefore only the latest—and certainly the most visible and intuitive—manifestation of a much longer evolution in which Artificial Intelligence has progressively become a decision-making infrastructure underpinning the main digital ecosystems.
Google Performance Max and Meta Advantage+: opportunities to seize
In recent years, the leading digital advertising platforms have profoundly changed the way companies can plan, manage and interpret advertising campaigns. In digital export, for example, businesses can increasingly rely on automated features within Google Ads and Meta Advertising to run initiatives aimed at driving online sales in international markets or generating leads from potential overseas business partners.
This shift, reflected in solutions such as Google Performance Max and Meta Advantage+, goes beyond the introduction of Artificial Intelligence. It represents a more fundamental transformation in decision-making roles: planning has moved from models requiring almost hands-on manual control towards increasingly automated systems. Companies are now primarily asked to define their business objectives, advertising budgets and creative assets, while algorithms take care of campaign execution and performance optimisation.
Under the traditional, more “manual” model—which is still widely used—most operational responsibility lies with the advertiser or the agency supporting the company. The business decides which channels to activate, how to allocate the budget across placements—on Google, for example, between Search and Display—which audiences to target, which countries to cover and even which times of day to prioritise.
This approach offers a strong sense of control, but it requires considerable expertise, in-depth knowledge of local markets and continuous campaign management. In international web marketing, this often results in highly complex structures: separate accounts for individual countries, dozens of campaigns running in parallel, localised creative assets and a significant investment of time in manually analysing performance and making adjustments.
Automated solutions such as Google Performance Max and Meta Advantage+ are designed to overcome these limitations. Instead of requiring companies or agencies to make hundreds of operational micro-decisions, they require fewer—but far more important—strategic choices: what is the business objective? What is a conversion worth? Which creative assets best represent the brand? And which data can help the system understand what truly matters?
The rest is entrusted to the algorithm, which learns from users’ actual behaviour and dynamically reallocates the budget towards the areas where it identifies the greatest likelihood of achieving the desired result.
Within the Google ecosystem, this change of paradigm is particularly evident in solutions such as Google Performance Max. Here, the very concept of a channel-specific campaign becomes less relevant. Instead of deciding exactly how much to invest in Search, Display or YouTube, the company defines a clear objective—such as selling a product or generating a qualified lead—and provides a set of creative assets, a product feed and reliable conversion signals.
The system then determines how to allocate the budget and which markets or contexts to prioritise. For companies operating in digital export, this means being able to activate campaigns across several countries at the same time without having to build and maintain complex local structures, allowing the algorithm to learn where demand is more responsive and where acquisition costs are more sustainable.
Meta Advantage+ follows a similar approach. Here too, the focus shifts away from manual targeting towards the system’s ability to identify the most responsive audiences autonomously. Advertisers no longer need to define interests, behaviours or geographic segments in a rigid way, but can instead focus on setting an objective, a budget and the relevant target markets.
For many Italian companies looking to sell internationally, this can provide a significant strategic advantage: new countries can be tested without requiring extensive prior knowledge of local audiences. By observing conversions and quality signals, the algorithm learns to identify common patterns among users in different countries and progressively shifts the budget towards the areas where it identifies greater efficiency.
What about LinkedIn?
This transformation is also taking place in B2B marketing. LinkedIn is progressively reducing the emphasis on the manual selection of job roles, industries and seniority levels in favour of campaigns focused on conversion objectives or qualified lead generation.
In an international context, this approach can help overcome some of the limitations of traditional demographic filters. A decision-maker in Germany and one in Spain may have different job titles but behave in similar ways throughout the buying journey. Automation can identify these behavioural similarities more effectively than purely manual segmentation.
To understand the practical difference between manual and automated approaches in international web marketing, imagine an Italian company manufacturing components for the food industry and looking to sell in Germany, France and Spain.
Under a manual model, the company would need to create three separate accounts or campaign sets, select different keywords for each country, allocate separate budgets, adapt creative assets and landing pages, and monitor performance on a daily basis to correct any inefficiencies. Every adjustment would require human intervention and a local interpretation of the data.
With an automated approach, by contrast, the company can create a single system focused on generating qualified leads and provide the platform with a limited number of key inputs: a properly tracked conversion objective, such as a request for quotation; a coherent set of multilingual creative assets; an overall budget; and an indication of the target markets.
The system then learns which combinations of message, country, context and audience generate higher-quality leads and progressively reallocates resources accordingly. The company does not lose strategic control, but deliberately gives up a degree of micro-level operational control.
Google Performance Max and Meta Advantage+: points to consider
However, this model does not work automatically in every situation. The first critical factor is time. Automated platforms need data, volume and continuity in order to learn. Short, intermittent campaigns or campaigns that are constantly modified risk interrupting the learning process just as it begins to become effective.
In international web marketing, this means accepting that the initial phase will partly involve exploration and that the strongest results may emerge over the medium term.
The second key factor is the quality of the objective. Automation works best when the goal is clear and measurable, such as an e-commerce sale or a clearly defined lead. When the objective is generic—for example, purely related to brand awareness—the algorithm may optimise for weaker signals such as views or interactions, which do not always translate into genuine business value, particularly in international markets.
Finally, there is the issue of brand safety, which is often underestimated. Automated platforms decide not only who sees an advertising message, but also the contexts in which it appears. For Italian companies whose positioning is built around quality and reliability, giving the algorithm complete freedom may create the risk of appearing in environments that are not fully aligned with the brand from an editorial or values perspective.
Automation should therefore not simply be accepted passively, but actively governed through rules, exclusions and constant monitoring of ad placement environments.
In this scenario, the role of companies and their consultants changes significantly. The task is no longer simply to “run campaigns”, but to design a system capable of providing algorithms with the right inputs: clearly defined business objectives, reliable tracking, consistent creative assets and high-quality data.
Automation does not eliminate strategy; on the contrary, it makes strategy even more important. It rewards companies that have a clear understanding of their business model and penalises those that rely on tactical shortcuts.