ECommerce Campaign Optimization: Display Ads and Search Keywords: 6-Wk Implementation

With ECommerce sites generating 15-20% sales for CPGs and this number seeing a continually increasing trend, organizations spend a portion of their marketing budget on running different campaigns on the ECommerce platforms.

Display Campaign Optimization:

Display campaigns featuring display advertisements are featured on several ECommerce platforms and paid media sites. They are an important tool for CPGs to build brand awareness, engage consumers, drive sales and gather valuable insights. But maximizing the returns obtained from these campaigns involves selecting the appropriate audience groups, campaign settings and budget allocated to individual campaigns, optimizing selected metrics such as ROAS, Total sales, Market Share, etc. The complexities revolve around answering the key questions:

1. Who is the right audience to target
2. What is the right budget against an audience group
3. How to optimize spend
4. How to determine the bid, frequency, etc. for a given context
5. When to change campaign parameters

Sigmoid has developed an ML based solution, in accordance to the Well Architected Framework of Azure, for optimization of all the above mentioned factors, relying on historical data for predictions and adjusting based on actual performance. Flexible to run either with or without 1P data, our solution has the capability to find the right audiences, bid values and other settings parameters by itself for a certain campaign category, automating the entire process and setting up the campaigns through DSP APIs. It allows the user to setup campaigns for any selected conversion metric, allowing user selected audience groups to be integrated as well. The continuous monitoring & optimization setup can stop underperforming campaigns and replace with a better setup as well.

Search Campaign Optimization:

Improving the category and positioning of the products on the search page of the ECommerce sites are of vital importance. With a part of the ECommerce Marketing budget being allotted in the bidding of keywords associated with every product, it becomes essential to improve the search rank of the products, in turn translating to incremental online sale. The following questions need to be answered:

1. How are products performing wrt vital keywords
2. What are my competitors leveraging
3. What other information from the product listing page of competitors can be put to use
4. What other key parameters can be used to improve the search rank

The solution built by Sigmoid, in accordance to Azure's Well Architected Framework, consists of a measurement and recommendation system. Measuring the performance of the current keywords, it also performs an inverse correlation analysis of competitors’ product performance, suggesting new keywords for better campaign performance. The solution also gives recommendations for bid-price optimization for the search terms which gets automatically ingested in the customer’s bidding platform as well. The frequency of the number of runs of the model can be decided by the customer.

The following Azure workloads have been used in developing the above mentioned solutions:

1. Azure Data Factory (ADF) for orchestrating data integration pipelines to integrate data from diverse sources
2. Azure Data Lake Storage, serving as the underlying storage layer
3. Microsoft Purview for a Unified Data Governance
4. Azure Machine Learning
5. Integrating and analysing large data sets
6. Power BI for Business Intelligence