- About Us
- Data Services
- Maintenance & Monitoring
- Consulting Services
- Customer Cases
- Blogs & news
One of the largest and most admired fashion and lifestyle companies in the world. The company operates within the online direct-to-consumer and retail business in EMEA. A strategic goal of the group is to put the customer at the heart of the business using the latest technology and data innovations to create the best customer experience. Their focus is to develop cutting-edge strategies around customer centricity, using advanced tools to drive customer growth with a long-term focus on value.
In order to achieve the goals for customer centricity, customer growth, and long-term focus on value, the company opted to leverage the functionality of Google Marketing Platform together with Google Cloud Platform to test and implement strategies that would allow for improved targeting, ease of activation, and insights on marketing performance.
With significant budgets going into paid search, the company team wanted to steer as closely as possible to the return on ad spend, not only on revenue but on profit. Taking profit into the calculation to find the optimal amount to bid ensures that the company would reach the highest quality visitors.
The Company Marketing Cloud integrates their product, order, browsing, and advertising data for insights, machine learning, and activation across Google Marketing Platform and Salesforce. Using the rich functionality of the GCP components, the company is able to design solutions that positively impact KPI’s through more precise targeting, seamless activation, and post-analysis capabilities to improve understanding of visitors and campaigns.
The overall concept is that the Smart Bidding algorithm learns to optimize on higher profitable transactions, and thus higher-quality customers would be prioritized for bidding.
Configuring the solution Crystalloids calculated conversion profitability in BigQuery and sent this to Google Ads for bid optimization in near real-time via Google Cloud Functions. The process was easily extended to new markets and brands.
Propensity models for predicting visitor engagement (category of interest, depth of visit, and source of visit) and intent to purchase, as well as the value of purchase, were constructed, where specific audiences were created for targeting as part of Search and Display initiatives. The process of developing these propensity models involved creating and maintaining a common, reusable feature store for models, constructing the models in BigQuery using Big Query ML, and automating the process for scoring the models in near real-time.
Within GCP, the training and monitoring of these models are also automated to ensure continuous tracking and refreshing of the models to reflect the most current environment at the company, thus ensuring relevance and value (see image above).
Automated steps include:
Google Cloud Platform
Google Marketing Platform
Share this story