Dynamic Video Ads Help Tazz Customers Find Local Dining Gems

Tazz is a Romanian food services platform that helps to connect restaurants with hungry customers. As the Romanian answer to Uber Eats, The Tazz team aims to showcase unique local dishes.

The company wanted to help customers locate restaurants that were offering famous international dishes with a local twist. To do this, they launched a marketing campaign involving partners from across Romania, each of whom had created their own unique dish.

However, Tazz still needed to find a way to effectively tailor their message to individual users local users based on their preferences. That’s where RTB House stepped in.

Tazz already had a brilliant video that encouraged local specific, but they wanted to go a step further. The company devised a plan to deliver relevant local content to users in each of the 28 Romanian cities that they are active in.

This type of campaign is significantly more complicated to pull off. It involves interpreting user data to select relevant content and offers from hundreds of potential local partners. Done manually, this could require hundreds or thousands individual creatives showcasing the dish of each partner.

It was also important to create a tailored approach to ensure that content was reaching the correct users. This did not just mean tailored content to geography, but also to the predicted food preference of each user. If this had to be completed by hand, this would be a daunting task.

Tazz wanted to ensure that the campaign was successful by focusing upon reach, viewability, and video completion rate (VCR).

To solve these challenges, RTB House took advantage of its state-of-the-art Deep Learning technology. The first step was to understand user groups, and create a way to dynamically display content that was likely to convert to individuals based on their location and preferences.

Once these groups were ready, the RTB House team created high-quality video ads that were able to substitute small clips of specific footage and offers depending upon the user in the form of video ads, and banner video ads. For example, each clip could showcase one of the individual dishes created by Tazz partners. This helped to increase the chance that a user would connect with this video, maximizing campaign efficiency.

Deep Learning algorithms were used to optimize the campaign over its course, ensuring that results continued to improve over time. This empowered Tazz to connect with users, and helped RTB House achieve some impressive results.

RTB House was able to beat a number of key metrics for the campaign. Deep Learning algorithms ensured that the viewability rate was 10.33% above the agreed benchmark. Personalized creative helped to create engaging videos, and a VCR that was 6.39% above expectations.

The team was also able to achieve some impressive results above and beyond the agreed metrics. The Click-Through Rate (CTR) for the campaign also exceed the results of other providers by more than three times.

The Tazz campaign in general helps to demonstrate the power of combining tailored messaging with cutting edge Deep Learning algorithms.