AI use cases & demos

Media Plan Performance Analytics Platform

by Max Frolov
CEO @ DataRoot Labs
Tech Stack:
Apache Spark
Ad tech
Project Length:
6 months
Read independent verified review on


  • A leading publicly traded Japanese advertising holding requested to build a media plan performance analytics platform. Tracking metrics provide invaluable insights to campaigns performance which ultimately allow for a better service to the customers, driving enormous value to the company and its shareholders.
  • The platform had to manage ads in order to understand campaigns’ performance using various data sources.
  • Our team was challenged with building the platform over the huge amounts of data streaming pipelines and required scalable, high-load implementations to achieve the goal.
  • Along with metrics monitoring module our team have developer campaign metrics forecasting module based on which valuable campaign-related decisions are made.

Tech Challenge

  • Solution had to be scalable to petabytes of data with stateless and stateful data points.
  • Processing of huge stream of data with tf.Datasets for further pushing it to the model which is responsible for metrics forecasting.
  • Configuration tf.FIFOQueue to efficiently pass train and test data into the model.
  • Model optimization with TensorRT which doesn't support numerous operations.


  • Our team has created a whole stack of technologies and corresponding dashboards to visualize the analyzed data and current campaigns performance.
  • Platform was built as a scalable micro-services solution with the ability to process data from hundreds of campaigns successively, asynchronously and in parallel.
  • Additional work was done to handle stateless and stateful data points in streams.


  • We have delivered a complex media plan performance analysis platform for ads management service.
  • Implementation allowed our client to get significant campaign quality improvements, resulting in a more tailored communication strategy, delivering more fitting ads for customers and increasing the bottomline revenue results.

Updated Jul 02, 2019 — 00:00 UTC