Goals
- Increase airline ticket sales using artificial intelligence and first-party data, without increasing the budget
- Raise the advertising conversion rate to 14% (from a baseline of 12%)
- Reduce the cost per acquisition (CPA) to €1.90 per sold ticket
- Implement an autonomous AI model with reinforcement learning that optimizes key campaign parameters in real time.
SPAR Slovenia interacts with millions of consumers every day through its physical stores, mobile app, online store, and the SPAR Plus Card. Although SPAR Slovenia had a large volume of first-party data, it had not effectively integrated this data into an operational advertising system until recently. The key challenge became identifying a user’s needs in real time and delivering a relevant ad, even if the user is not integrated into the CRM.
Project objectives
Spar set itself the goal of creating an architecture that delivers the following:
- Microsegmentation at the level of the individual
- Real-time ad creation and display
- Effective targeting of related users (lookalikes)
- Long-term data sovereignty, independent of global advertising platforms
Solution: The Tailormade architecture
In cooperation with iPROM, SPAR Slovenia developed Tailormade, a data infrastructure built on the iPROM Private DMP platform. It integrates data from all touchpoints – including POS systems, the web, and email – into a single intelligent core. The system operates on three levels:
- Microsegmentation: users are segmented based on behavioural patterns rather than demographics. Each profile is continuously updated based on current behaviour.
- Micro-targeting and content creation: the AI automatically generates an ad for each user, including text, visuals, and a call to action. These ads are displayed in a dynamic format iPROM Spider Ad on premium Slovenian online media properties at the right moment.
- Lookalike Audience Modelling: Algorithms identify users who resemble SPAR Slovenia’s best customers and deliver tailored and effective content.
Tailormade has already proven its effectiveness within its first three months of operation:
- 85% higher behavioural accuracy in matching content to user interests.
- Over 100,000 new users reached outside the CRM database.
- 45% higher CTR (from 0.49% to 0.71%) for behaviourally targeted ads
- 30% lower media bounce – more impressions delivered to users interested in the content.
- 100% of ad impressions were based on first-party data, which means full control over the data chain.
Measurement and further development
The system is designed to be measured and optimized in real time. Each user interaction generates data that is integrated into the behavioral model. Tailormade is not just a campaign; it’s a smart infrastructure that constantly learns and makes advertising decisions independently.
Vision 2026
SPAR Slovenia plans to expand Tailormade by integrating generative AI for content creation and predictive models by product category. The goal is to have more than 100,000 in-depth behavioral profiles to enable even greater personalization and higher conversion rates.
Conclusion
Tailormade represents a shift away from traditional approaches to digital advertising. Rather than optimizing campaigns, it focuses on developing an ecosystem in which brands use AI to understand users, learn from their behavior, and present them with relevant messages at the right moment. With Tailormade, SPAR Slovenia is shaping the future of data-driven communication.
“In collaboration with iPROM, SPAR Slovenia developed a marketing system that uses our data to enable precise, behaviorally tailored advertising. With Tailormade, we connected data from all touchpoints for the first time and used AI to create ads served to the right user at the right moment. This has taken personalization to a new level, expanded our reach to users outside the CRM system, and created a comprehensive, data-driven marketing ecosystem,” explained Aleš Popič, Head of Marketing Communications at Spar Slovenia.