| Title |
Marketing campaign performance summary |
| Authors |
Šukytė, Auksė ; Gumuliauskaitė, Vaiva |
| DOI |
10.15388/Gronskis.2026 |
| Full Text |
|
| Is Part of |
20h Prof. Vladas Gronskas international scientific conference, 8th of May 2026, Kaunas, Lithuania : abstract book.. Vilnius : Vilniaus universiteto leidykla. 2026, p. 52-53 |
| Keywords [eng] |
BigQuery ML ; logistic regression ; K-means clustering ; marketing campaign performance |
| Abstract [eng] |
Marketing performance analysis is essential for improving campaign effectiveness. The research aims to analyze and predict marketing campaign performance using machine learning in Google BigQuery. The results are based on a Marketing Campaign Performance dataset containing 10,000 records, sourced from Kaggle. Logistic Regression model enabled to classify successful campaigns based on impressions, clicks, cost and ROI, and has achieved an exceptional accuracy rate of 98%. The K-means clustering model is employed to segment channels into four distinct groups based on their ROI and cost-efficiency. The model identified Cluster 4 as the most significant group, containing biggest number of campaigns and highest return on investment (ROI). The findings reveal that influencers are primary driver of success, while the Search channel demonstrates the highest stability across timeframes. The results validate that combination of clustering and classification algorithms builds a robust framework for identifying high-impact marketing channels and valuating their performance. |
| Published |
Vilnius : Vilniaus universiteto leidykla |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2026 |
| CC license |
|
