From Revenue to Profit: Shoppster Increased Non-Brand Net Profit by 143%

Brand

Agency

Verticals

E-commerce & Retail

Geography

Serbia

About Shoppster

Shoppster is a leading regional e-commerce platform operating in Serbia and Slovenia. Designed as a digital one-stop shop, it offers customers a fast and seamless online shopping experience, with more than 800,000 products available in Serbia and 300,000 in Slovenia.

Alongside flexible payment options, a loyalty programme and educational video reviews, Shoppster also supports sellers through reliable logistics and direct access to a customer base of more than two million potential buyers.

Challenge

Revenue alone was not giving Shoppster a complete view of campaign performance. While Google Ads could see transaction value, there was no profit signal in the account to reflect the actual profit remaining after margins, cost of goods sold, and fulfilment expenses. By relying on raw revenue, Smart Bidding blindly scaled low-margin products.

At the same time, heavy brand-search noise distorted every non-brand category’s apparent return. This brand leakage meant that branded searches were inflating non-brand performance baselines by up to 28%, making it harder to identify which products and categories were truly driving growth.

Shoppster’s goal was to make profitability the core signal behind bidding and budget decisions by feeding clean, verified first-party profit data back into Google Ads in real time.

Key Objectives:

Use transaction-level profit as the primary optimization signal.
Include margins and fulfilment costs in campaign measurement.
Separate non-brand performance from branded search influence.
Feed reliable first-party profit data into Google Ads in real time.
Shift budgets toward the most profitable products and categories.
Increase profit while maintaining efficient ad spend.

Missing profit data

Brand-search distortion

Low-margin products prioritised

Strategy

The Approach

With strategic support from Coob Ads and in collaboration with Escape and PopUp Advertising, Shoppster rebuilt its measurement setup so Google Ads could optimize toward actual profit rather than revenue alone.

Server-side profit measurement: A first-party pipeline using server-side Google Tag Manager and Google Cloud Firestore calculated a custom dynamic variable, “Purchase SST Profit”. This represented the net margin of each transaction based on live product margins, cost of goods sold and fulfilment expenses, feeding the data back to the ad platform in real time.
AI-powered brand filtering: Acting as an automated data-hygiene layer, a Google Gemini-assisted workflow interpreted thousands of search queries to identify and remove 152 brand-name variations and spelling errors from non-brand optimization data, creating a cleaner view of incremental performance.
Margin-based conversion values: Verified profit data was routed directly into the Google Ads Conversion API and used by Maximize Conversion Value, allowing Smart Bidding to prioritize transactions based on profitability.
Profit-driven budget allocation: With net margin as the optimization signal, Google’s AI was granted the autonomy to dynamically shift spend toward more profitable products and categories while reducing investment in inefficient segments.

Tools & Technologies

Google Tag Manager Conversion API Smart Bidding

Success

The Results

With profit-based optimization in place, Shoppster was able to scale campaigns based on actual commercial value rather than revenue alone. Non-brand net profit and POAS increased significantly, while budgets were directed more efficiently across product categories.

+143%

true non-brand net profit YoY, with a 33% increase in total ad spend

+83%

non-brand portfolio POAS, moving beyond the previous break-even baseline

+119%

net profit and +169% ex-brand POAS in Small Appliances, with 21% lower ad spend

+160%

non-brand net profit and +52% POAS in Large Appliances

"Before this project, our campaigns were optimizing toward revenue without fully understanding the profitability behind each transaction. By building a secure first-party data pipeline and introducing an AI-powered data-hygiene layer, we were able to remove distorted signals and feed actual profit data directly into Google Ads. This transformed our bidding strategy, enabling the system to prioritize genuinely profitable segments and allocate budgets more efficiently. We now have a much clearer, more commercially relevant foundation for sustainable growth."

Igor Možek

Head of performance