Lebara

Lebara is a mobile virtual network operator (MVNO) present in five European markets. Its mission is to offer customers a better value-for-money alternative to traditional telecom operators. In the French market, Lebara runs its media acquisition through Google Ads and several social platforms, with a strong need for reliable data to decide where to invest.

Challenge

With cookie banners, ad blockers and restrictive browsers like Safari, a growing share of Lebara’s data was becoming what we call dead-end data. These are real events (a visit, an add-to-cart, a conversion) that happen in the browser but whose signal stops along the way, before reaching media platforms or analytics tools. The data exists, but it is invisible to the systems that need it.

The problem is therefore not just a loss of volume. Bidding algorithms like Google Ads and Meta, deprived of part of the conversion signals, optimize on an incomplete picture of customer behavior. This puts advertising ROI at direct risk, along with the reliability of the reporting used to allocate budgets.

The challenge was twofold: recover part of this lost data and govern it better, while reducing the number of tracking scripts loaded on pages to speed up loading time, improve user experience and boost search engine visibility.

Lebara had already implemented Google Tag Gateway (GTG), a solution that routes Google tags through the site’s own domain rather than a third-party domain (known as “first-party”), which limits blocking. This first step remained limited in several ways:

• Coverage limited to Google tags only: tags from other media partners (Meta, TikTok, Snapchat, affiliate partners) kept loading the classic way, without benefiting from first-party.
• Limited improvements on Safari: the browser’s cookie restrictions (ITP) still limited the recognition of returning visitors.
• No web performance gain: GTG reduced neither page weight nor the number of requests loaded by the browser.
• Dependence on a single vendor: covering all media partners required a solution able to route every tag, whichever the vendor.

These limits illustrate the nature of the problem, GTG reduced dead-end data on the Google channel alone, but left other media partners in the same blind spot. Yet a bidding algorithm cannot work with a partial signal, and a degraded matching rate on Meta or TikTok translates directly into less accurate bidding and a higher CPA on those platforms, exactly as it did on Google before GTG.

Other indicators confirm this trend. Events captured per GA4 session are up 1.7 points since before GTG. Media platforms are also receiving more complete signals, with purchase events captured up 8.5% on TikTok and 9.7% on Snapchat compared to client-side tracking alone.

Bypassing Safari’s ITP restrictions, already started with GTG, was reinforced by Addingwell’s cookie restoration solution. The share of detected Safari users increased, while the rate of new visitors in this environment decreased, a sign of better recognition of returning visitors. The share of visits coming from Safari increased by 57%.

External factors may also contribute to these changes, which makes it impossible to isolate precisely the share attributable to SST, although it is certainly significant.

Conclusion

By moving from a first step limited to Google tags (GTG) to a server-side solution benefiting all media platforms with Addingwell, Lebara extended this first-party bypass to all its media partners, while reducing page weight and improving its matching indicators.

By eliminating dead-end data across all its media partners, and not just on Google, Lebara turned previously lost signals into data that bidding algorithms can use. That is precisely why the gains are not limited to Google Ads (-37% CPA). The increases in purchase events captured on TikTok (+8.5%) and Snapchat (+9.7%) show that the benefit of solving dead-end data materializes differently depending on the platform, but with the same underlying logic: more available signal, better optimization.

Some gains remain partially attributable to other factors, such as the team’s media optimizations or seasonal effects. Overall, though, the results confirm the value of centralized, first-party tracking for a high-volume telco player.