Consider Sara, a frequent traveller using a platform like Wego or Expedia. One week she is searching for a quick one-way hop from Dubai to Riyadh for a client meeting: schedule is everything; price is irrelevant. Six weeks later she is back, this time hunting for a ten-day family holiday to the Maldives, budget front of mind, and non-stop flights a must for the kids. She is the same customer with the same login, yet her preferences have shifted completely. Most brands just see two searches. Smarter brands see two entirely different needs and act accordingly.
That distinction is at the heart of a growing challenge for marketers across the region. As third-party cookies disappear, many companies are doubling down on the first-party data they already collect from their own websites and apps. The instinct is right. The execution, however, often falls short.
The end of third-party cookies is a wake-up call, not a crisis
For years, the dominant logic of digital marketing in the region has relied heavily on third-party signals: data purchased or rented from external platforms to build audience profiles and target ads. That era is closing. Global privacy changes, new data protection laws in Saudi Arabia and the UAE that came into force in recent years, and growing consumer awareness are all converging to cut off the external data supply.
The instinctive response from many firms has been to shore up their CRM systems, invest in email capture, and hope that purchase histories will fill the void. But this misses a critical insight: the most valuable signal your customers send you is not what they buy. It is everything they do before they buy.
The click trail: a goldmine most companies ignore
Every search query entered, every product clicked, every filter applied: together, these micro-interactions form what researchers call the customer journey, a trail of behavioural data collected entirely on your own platform and far richer than any transaction record.
Recent academic research illustrates just how rich this can be. Studying a major online travel platform, researchers found that just two clicks into a search session, a platform could predict a customer’s preferred airline with 25 per cent greater accuracy than at the start of the visit. After five clicks, accuracy improved by 73 per cent, and combining live click data with past journey behaviour predicted the eventual purchase ten times more accurately than systems relying on purchase history alone.
The data to understand your customers better is already on your servers. The gap is not in collection. It is in interpretation.
Why most firms still get this wrong
The fundamental mistake companies make is treating every customer visit as if it belongs to the same person with the same need. Sara booking an expensed business trip one month and funding a family holiday out of her own pocket the next is effectively a different shopper on each occasion. Her priorities, price sensitivity, and tolerance for stopovers shift with context.
Most marketing systems are not built to recognise this. They aggregate customer behaviour into a single profile and make predictions based on who this person has been, rather than what they need right now. The result is retargeting ads that show someone a budget airline option after they have spent twenty minutes filtering exclusively for non-stop premium flights. The data was there. The insight was not.
Our research found that customers naturally fall into distinct journey types: segments defined not by who they are but by what they need on a specific occasion. Among travel customers alone, we identified 22 such patterns, from last-minute solo domestic trips to family holidays booked months ahead. The same customer fell into very different ones on different visits.
When retargeting was guided by journey-type intelligence rather than simply featuring the last product a customer clicked on, click-through rates improved by 28 to 32 per cent. That is not a marginal gain. It is the difference between a retargeting campaign that pays for itself and one that does not.
What this means for personalisation in the region
For marketers across the Gulf and the wider Middle East, this matters now. Digital commerce in the region is maturing rapidly, and competition is intensifying. Companies that learn to read live first-party signals will have a structural advantage over those relying on purchase histories or vanishing third-party data.
There is also a cold-start dimension worth noting. Many regional consumers are still early in their relationship with certain platforms and lack purchase history. But they click from session one, and just five clicks is enough for behavioural patterns across customers to predict a new user’s preferences almost as accurately as a long-established one. In a privacy-first world, the goal is not knowing more about the customer as a person but understanding what they need right now.
Three things marketers can do
First, store the journey, not just the transaction. If your platform logs purchases but discards clickstream data, that architecture is working against you. Searches, dwell time, and the filters a customer applies are your clearest real-time signal of what they actually want.
Second, segment by occasion, not just by profile. On any given visit, the occasion often matters as much as the history. Use live behavioural signals to classify what kind of need this session represents before a purchase is even made.
Third, personalise the results page before you personalise the ad. The highest-leverage point is not the retargeting email; it is the results page the customer is actively browsing. Moving the product most likely to match a customer’s current preferences to the top of results reduces session abandonment and more than doubles click-through on that item. The data to do all of this is already there. The question is whether you are listening to it.
Nicolas Padilla is an Assistant Professor of Marketing at London Business School, and co-author of The Customer Journey as a Source of Information.
