ML models development for influencer marketing platform

As a digital marketplace connecting brands and influencers with impactful audiences worldwide, our client recognized the need to enhance the functionality of their influencer marketing platform. With the help of HMD LLC's big data and data science experts, they were able to achieve this goal through the implementation of a proprietary advanced social media analytics solution.

Prior to partnering with HMD LLC, our client had already established a successful online marketplace for top social media influencers to connect with brands. However, the existing analytics tool only fit Instagram analytics and had several limitations that prevented the company from further developing its software-as-a-service analytics part.

The main improvements our client sought to make were the ability to enable real-time analytics, expand to new social networks, and use machine learning for influencer analytics. HMD LLC was chosen as the expert partner due to its extensive experience in both data science and social media analytics. This reduced the time and cost of the project, making it a crucial factor in the decision to work with HMD LLC.

HMD LLC was tasked with building a two-sided platform, one side for influencers to access data such as follower gain or loss, audience demographics, and optimal posting times, and the other side for advertisers to set campaign goals and track reach through metrics like hashtag mentions, total views, and new followers.

To achieve this, HMD LLC built a solution that downloaded all publicly available data from social networks such as Instagram, TikTok, and Twitter. This data was then used to create BI reports or as input for predictive models that predicted the gender, age, location, and interests of social media users. The solution was built on Hadoop Stack, including HBase database and Kafka for real-time data pipelines, as well as utilizing Amazon Web Services such as EC2, Autoscaling, S3, RDS, SageMaker, Kinesis, and Glue.