Exclusive: AI-Driven Advertising Shifts Focus to Long-Term Customer Engagement

In the rapidly evolving world of advertising, the metrics of success are often defined by immediate conversions and short-term results. But as the market matures and technology advances, it’s essential to consider a deeper, more strategic approach to how advertising systems can drive real value for businesses. With Middle East e-commerce set to soar, projected to reach US$50 billion by 2025, are we truly maximising the value of our advertising platforms? As it turns out, the answer lies not in our traditional methods of predictive targeting but in a fundamental shift towards understanding the entire user journey and building systems that recommend the right product at the right moment, no matter where the user is in their decision-making process.

Let’s explore this shift in advertising philosophy and how AI and machine learning can enable a more sophisticated, long-term approach to advertising.

by Mira Weiser, Chief Product Officer at Yango Ads and Vasiliy Pushkin, Machine Learning Lead at Yango Ads

The Shift in Advertising Philosophy

Advertising systems have long operated as recommendation engines—designed to suggest products based on user behaviour and preferences. Yet advertisers have traditionally treated them as audience-targeting tools, focusing on who should see an ad based on predictions of their interests or likelihood to convert rather than what is being shown. While effective in many cases, this mindset limits the potential of ad systems to drive meaningful engagement.

The future of advertising lies in improving the product recommendation itself. Rather than just predicting which users are likely to convert, systems should identify the most relevant product experience for each user—whether or not they’re actively searching. This requires a shift from targeting personas to a hyper personalized experience by understanding individual contexts: interests, intent, and needs at the optimal moment.

The technology to do this is evolving, but unlocking this next phase demands deeper AI integration, real-time data processing, advanced machine learning models, and a nuanced understanding of the customer journey. It’s not just about predicting who will buy, but anticipating what will truly resonate—even if they aren’t aware of it yet. This evolution aligns advertising more closely with its original purpose: discovery, relevance, and value creation.

Ad Platforms as Recommendation Engines

Viewing advertising systems as recommendation engines reframes their role—not just as tools for driving conversions, but as sophisticated systems that predict and surface the next best offers for each user. Whether a user is actively searching or casually browsing, the system should intuitively suggest what best matches their evolving needs and present that offer in a way that feels seamless and intuitive. In fact, research shows nearly 90% of consumers prefer personalized ads, and 87% are more likely to click on ads for products they’re interested in or shopping for, so this method would prove effective.

But realizing this vision comes with a major challenge: scale. To recommend the most relevant product for every user, the platform must evaluate a vast product catalogue in real time—a task that’s both data-intensive and costly. As a result, most systems default to promoting familiar, high-conversion items with proven success, optimizing for short-term conversions.

This approach, however, creates two key problems. First, advertisers end up paying for users already close to converting—those who would likely buy anyway—leading to inflated costs with little added value. Second, it sidelines long-tail products, which, despite lower visibility, can account for up to 60% of Gross Merchandise Value (GMV) in some sectors.

The real problem is data sparsity. Long-tail products and long-term interests don’t appear frequently in interaction data, making them harder to surface with traditional models. To address this, ad platforms must adopt advanced AI techniques—like reinforcement learning, semantic IDs, and large language models—to dynamically adapt recommendations abased on what is more likely to provide value to the user, even if it is not as popular.

By doing so, advertisers can move beyond short-term conversion tactics and unlock broader demand, tapping into overlooked opportunities that are powerful drivers of GMV.

Demand Generation, Not Just Conversions

In an ideal world, advertising systems wouldn’t just chase conversions—they’d actively generate demand. Rather than targeting users already primed to buy, platforms should help users explore broader interests, guiding them from discovery to conversion through a deeper understanding of the full customer lifecycle.

This marks a true paradigm shift. It’s no longer about showing familiar products and hoping for a quick win. It’s about anticipating future interests and nurturing curiosity into intent. With deep learning models that analyze a user’s complete interaction history, ad systems can predict evolving needs and deliver ads that shape long-term engagement, not just immediate action.

Tracking LTV to Measure Success

To truly benefit from this shift, advertisers must move beyond short-term metrics like cost per acquisition (CPA) or return on ad spend (ROAS) and focus on the Lifetime Value (LTV) of new users. Many campaigns target new users but fail to track their long-term contribution. Without measuring LTV, marketers miss the real ROI of their ad spend and underestimate the true cost of attracting a new user into their ecosystem.

Tracking LTV reveals how much a new user will contribute to your bottom line over time, enabling smarter investment decisions. Once LTV is understood, the metrics can be incorporated into ad systems. With AI,

advertisers can identify high-LTV segments and optimize future campaigns to focus on long-term growth, not just immediate conversions.

What Advertisers Should Prioritize

While the ideal system isn’t here yet, advertisers can take practical steps toward a more sustainable, long-term strategy. Start by providing ad platforms with the full product catalogue to improve recommendation relevance – the more data, the better. Align marketing KPIs with business goals by tracking the entire customer journey and measuring user lifetime value from users across channels, enabling smarter ad spend. Distinguish the true cost and value of acquiring new versus returning users to fine-tune ad strategies to target the most profitable audiences. Finally, track user engagement throughout the funnel, from discovery through conversion, to better understand and optimize long-term demand generation. The tools to track LTV are already here. We just need to shift our mindset and move beyond short-term performance metrics to measure what truly matters in the long run.

Moving Toward the Ideal

Some of the largest advertisers are beginning to move beyond conversion-focused strategies, shifting toward traffic generation to maximize LTV over time. By bringing users into their ecosystem early, they rely on the system to nurture long-term value—a notable change in how ad performance is measured.

As this mindset spreads, advertising will evolve into a more sustainable, value-driven model. The focus will shift from short-term wins to building lasting user relationships. With AI and machine learning advancing, advertisers have a powerful opportunity to drive demand, optimize for LTV, and unlock meaningful, long-term ROI.