Development of a Marketing Bot for Social Media Management
Objective
For the eCommerce company PosterXXL, we developed a social media management bot capable of automatically managing and optimizing ad budgets.
To power this automation, we used a custom-built machine learning algorithm, continuously trained on real-time sales data.
The goal of the project was to reduce customer acquisition costs in the short, medium, and long term, while also simplifying the management of paid social media campaigns.
About Our Case Study Partner
For nearly two decades, PosterXXL has been dedicated to turning cherished memories into high-quality photo products. They offer a wide range of premium items for every budget, along with user-friendly tools for easy customization and design.
Meta Ad Account Management
Initially, we were tasked with finding a short-term solution for managing the Meta ad account while continuing to improve overall campaign performance.
To achieve this, we implemented our proprietary machine learning algorithm, which was able to identify optimization opportunities in real time.
In a second phase, the logic was applied in a daily routine, enabling ongoing improvements. This led to a significant reduction in customer acquisition costs and greatly reduced the need for manual account management.
Algorithm Development
Based on a custom cost-benefit framework, we developed a reinforcement learning algorithm specifically trained to optimize the cost-per-purchase (CPP).
The biggest challenge was creating a reliable and meaningful cost function to deliver accurate positive or negative feedback to the system.
This foundational model has since been successfully applied in numerous other projects.
Bot Development
The bot was developed using Golang and was powered by data retrieved via the Meta API. Advertising account data was stored in our database using scheduled CronJobs, executed via AWS infrastructure.
In the next step, we applied our standard machine learning model to the available data in order to identify optimization opportunities.
After manually reviewing the results and applying selected changes, we waited for positive validation of those adjustments. Once confirmed, the system was extended to support daily automated updates.
A key element was the ability to intentionally make no changes, to avoid confusing the Meta algorithm with too many minor adjustments — a crucial insight for stable performance.
Despite automation, the system continued to undergo regular manual review, especially to improve edge cases and ensure overall ad account performance remained strong.
Let’s build the right social media strategy for you
Why Us?
Broad Tech Stack
Depending on your database system, as well as your backend and frontend setup, we can develop solutions that are tailored to your unique requirements. We have the expertise to build custom solutions using a wide variety of programming languages and technologies.
API Expertise
We have in-depth experience working with a wide range of APIs — especially those from Google and Meta — and have already implemented numerous custom solutions.
In principle, we can integrate and utilize almost any API, adapting it precisely to your system and goals.
Ad Account Optimization
With our many years of experience — including as a certified Google Ads Partner — we can identify weak points in your ad accounts and optimize them based on your specific goals. Our approach is data-driven, strategic, and always tailored to your business needs