With Sellergini’s GrowZ products, our clients can track their AdSpend and Sales by the hour and determine the most effective time to target their ads. We have helped our clients reduce their Amazon advertising cost of sales (ACoS) and set goals to manage Amazon ads.
How does it help?
The key to reducing advertising costs is choosing the right time to target ads. We have tried several machine learning techniques to predict the most opportune time for our clients to send Amazon Ads to their customers. The hybridized Genetic Algorithm, which was published in the International Journal of Computer Applications, has helped our clients achieve remarkable results. The study uses a novel approach to find a global optimum by using operators from the Employed Bee Phase of the Artificial Bee Colony to run Genetic Algorithms.
For finding the appropriate Amazon Ads time, we explored other methods such as neural networks, anomaly detection, and clustering methods. Compared to the other approaches we have tried, the hybrid Genetic Algorithm-based approach improved ACoS for our clients by 2x-3x. We even ran A/B testing experiments where we divided Amazon Ads targeting between hybridized genetic algorithms and neural networks. The hybridized Genetic Algorithm approach reduced ACoS by 55%, whereas a neural network-based approach only reduced ACoS by 26%.
We continue to explore such innovative research in the field of AI at Sellergini to boost the performance of our clients’ Amazon Ads campaigns. By using this hybridized Genetic Algorithm approach, we have been able to reduce the costs of ads for our clients, and we recommend it to others.
 Singh, Deepak, and Ankit Sirmorya. “Solving Real Optimization Problem Using Genetic Algorithm with Employed Bee (GAEB).” International Journal of Computer Applications 42.11 (2012): 1-5.