Performance Marketing

Performance marketing is the set of advertising activities in which optimisation and measurement are tied to concrete results: clicks, leads, sales, acquisitions. At Bliss we manage it as a system architecture, not as isolated individual campaigns, integrating Google Ads, Meta Ads and TikTok Ads on a single tracking, analysis and reporting infrastructure. The principle is that every euro invested must produce data, and all data must feed a decision.

The structural difference lies in the type of demand they capture. Google Ads captures explicit demand: the user is actively searching for something and the system meets them at the moment of highest commercial intent. Meta Ads works on latent demand: it reaches the user before they have formed an explicit search intent. In the systems Bliss governs, the two channels are used in a complementary way: Google for direct conversion, Meta for audience building and structured retargeting.

TikTok Ads deliver measurable results in highly visual sectors and in under-35 demographic segments. Their effectiveness depends on the ability to produce content native to the platform: content designed for other channels and then adapted does not work. At Bliss we always assess whether to bring TikTok into the system after analysing the target audience and the partner's production capacity. It is not a platform choice: it is a choice of editorial architecture.

The minimum budget needed to produce statistically significant data depends on the sector, the average cost per click and the target cost per acquisition. At Bliss we set the threshold after the initial audit: it cannot be done beforehand without analysing data for the specific market. As a general guide, budgets that are too low do not allow platform algorithms to gather enough conversion volume to optimise: campaigns remain in a permanent learning phase that never produces scaling.

ROAS (Return on Ad Spend) is the ratio between the revenue generated by campaigns and the advertising spend invested. A ROAS of 4 means that every euro spent on advertising generates 4 in revenue. At Bliss we always measure net ROAS as well, adjusted for product cost and margin, because gross ROAS can be misleading if not set against the partner's business model. The goal is not a high ROAS in absolute terms: it is a ROAS that is sustainable for the specific cost structure.

Conversion tracking is the infrastructure that makes it possible to attribute every sale or lead to the campaign that generated it. Bliss implements it through Google Tag Manager, the Meta pixel, Google Ads tags and GCLID integration to close the loop between the campaign and the partner's CRM. Where possible, we integrate Meta's Conversions API to reduce the signal loss caused by browsers with ITP enabled. Without reliable tracking, campaign optimisation works on data that does not reflect reality.

The first significant optimisations appear between 4 and 8 weeks after launch, the time the platforms' algorithms need to gather sufficient data. Structural results — those that justify scaling decisions — emerge between 3 and 6 months. At Bliss we do not set timelines before the analysis: they depend on the sector, the level of competition, the quality of the initial tracking and the volume of conversions the business is able to generate.

Yes, through attribution modelling techniques and, where possible, by integrating Meta's Conversions API, which allows offline conversion events to be sent directly to the platform. At Bliss we assess case by case which attribution model is most accurate for the partner's business, avoiding overstating digital channels in reports or underestimating their real contribution to revenue.

On three distinct levels. The first is campaign ROAS: how much revenue each euro invested generates. The second is CPA: how much it costs to acquire a customer. The third is the LTV/CPA ratio: the long-term value of the acquired customer relative to the cost of acquiring them. At Bliss we define the reference metrics before starting: they are not chosen after the fact based on what performed best, but set during the architecture phase as binding operational targets.