How did we decrease conversion cost by 93% for a client in home&garden industry?

The client is a global manufacturer of garden and patio equipment, actively operating in 19 markets—across Europe, the USA, Australia, the developed economies of Asia, New Zealand, the United Arab Emirates, and Central America.

Its offer includes over 1,000 products, all bearing its own brand. The target audience consists of people who value spending time outdoors and appreciate the highest quality of garden and patio equipment. The client’s products belong to the premium segment—they are durable, solid, and covered by a multi-year warranty. A characteristic feature of the client’s business is seasonality—the most intensive marketing and sales period falls between March and July.

We wanted to start our activities as quickly as possible. That is why, immediately after the decision was made to launch the marketing activities, we began working with all the key departments on the client’s side—regional managers in Europe and the analytics team in the USA. To ensure maximum control and full transparency, we agreed on a schedule of regular meetings right from the start.

The initial stage was not the easiest—compared to established foreign markets, Poland performed considerably worse, particularly in terms of e-commerce results across all channels, not just DV360. However, the client, with whom we have worked for years, was aware that entering a new market is a process that takes time. They understood perfectly well that every market has its own specifics and dynamics.

A strategy based on data and branding campaigns

In the first months, we focused on collecting and analyzing data and identifying competitive advantages. We searched for the most effective traffic sources, precisely defined the target audience, and built brand recognition among potential customers.
The main premise of our programmatic strategy was to make maximum use of the budget on awareness-building campaigns—generating as much website traffic as possible and increasing brand recognition and familiarity. As part of these activities, we launched:

• Seasonal campaigns—spring season kickoff promotions, including an offer with a free gift for people who purchased the client’s product (not only in the online store).
• Display campaigns—based on both Google audience lists and 1st party data, as well as machine learning mechanisms from Google Analytics 360, taking into account not only clicks but also impressions.
• Masthead—reach campaigns on the YouTube homepage, configured directly from the Display & Video 360 panel.
• Video campaigns—run on YouTube, based on remarketing lists from the Masthead campaigns, data from GA360, and Google audience lists, aimed at strengthening brand recognition.
• Product campaigns—dynamic remarketing using a feed from Google Studio and remarketing lists generated as part of earlier branding activities.
• Private Deals—video campaigns run on the Spotify platform.

After just two months, the branding activities began to bring satisfying results—including in the online store’s performance.

After the first few months, a growing bounce rate and a declining number of post-click sessions became a challenge—a phenomenon typical after entering new markets in programmatic activities. We managed to limit this effect thanks to better audience targeting, based on data collected in GA 360 (mainly view-through conversions, which for programmatic are one of the most reliable data sources).

Implementing a split of KPIs and optimization across prospecting and remarketing campaigns

Based on data about website and online store users, we began creating separate strategies for two segments: new traffic and remarketing. We focused not only on the appropriate allocation of budget, but also on close cooperation with the client’s analytics department to develop personalized bidding algorithms.

• Implementing a split of KPIs and optimization across prospecting and remarketing campaigns — based on data about website and online store users, we began creating separate strategies for two segments: new traffic and remarketing. We focused not only on the appropriate allocation of budget, but also on close cooperation with the client’s analytics department to develop personalized bidding algorithms.
• Implementing custom bidding algorithms — we defined micro-conversions and assigned them values, calculated based on actual consumer behavior. We developed a script whose goal was to maximize ROAS among users visiting the online store. In parallel, we also created a second script, focused on minimizing the cost per session in prospecting campaigns.
• Brand Lift Surveys — monthly studies on YouTube analyzing the impact of ads on brand recognition and competitive advantage. As part of these activities, we conducted regular Brand Lift surveys available on YouTube, aimed at analyzing the impact of video campaigns on brand recognition and its competitive advantage.

Thanks to appropriate budgets and high-quality video materials, we were able to carry out studies that were particularly important in the initial phase of the branding activities. At this stage, the data came not only from tools measuring website traffic, but also from market analyses and competitive benchmarks. This allowed us to precisely identify the most effective video materials, the best-converting audience lists, and the demographic groups in which the brand achieved the highest level of recognition.

Programmatic activities in Poland in 2021 were run from February to August. During that time, we gathered enough data to start the new season in March 2022 fully prepared—drawing on the infrastructure built the previous year, ready-made audience lists, and appropriately developed ad materials, campaign types, and personalized bidding algorithms for prospecting and remarketing activities.

This allowed us to precisely define key performance indicators (KPIs) and implement optimization for ROAS and cost per session.

The results we achieved exceeded our expectations—after just the first three months of the programmatic campaign, we managed to reach all the goals we had set, while at the same time reducing the budget by 60% compared to the same period the previous year.

The effects are best illustrated by data from Google Analytics 360 in the e-commerce section relating to the DV360 platform. A comparison of the first three months of the 2021/2022 season (March–May) showed:

• a 57% reduction in cost per session,
• a 116.08% increase in ROAS,
• a 93% reduction in cost per conversion.

Such significant results were achieved primarily thanks to the implementation of custom bidding algorithms. During the off-season (September–February), we did not carry out any activities other than dynamic remarketing in DV360. We gathered all our conclusions and insights during the six-month campaign cycle in 2021.

Entering a new market without prior data required flexibility, patience, and a strong analytical foundation. Focusing on brand-building activities based on upper-funnel data allowed us to create a solid foundation for further growth. Thanks to automation, a precise strategy, and continuous testing of new formats, we achieved real results, even though the main goal was not online sales but supporting the distributor network.

The trust the client placed in us resulted in a long-term collaboration that we continue to this day. A partnership-based approach, effective activities, and transparent communication became the foundation for further scaling our joint projects.