Young Mavericks’ Machine Learning Tool Optimizes Social Blue’s Online Campaigns

There is a good chance that you have come across one of Social Blue’s advertisements while scrolling through your timeline. Their expertise in lead generation multiplies followers, website visitors and donors through online campaigns for numerous domestic and international companies. Social Blue requested Young Mavericks’ knack for data to help explore the possibilities for optimizing their campaign texts through machine learning. Our Proof of Concept sought to answer the company’s question: ‘Can machine learning improve our campaign results based on previous ad performances?’

Proof of Concept: ‘Previewing’ Data Innovation

A Proof of Concept provides Young Mavericks’ new partners the chance to gain initial insights and start the transforming their work into data-driven processes. This ‘preview’ defined a project with clear objectives and exploited Social Blue’s endless possibilities for data innovation. After agreeing upon the Proof of Concepts’ parameters, a team from Social Blue, together with Young Mavericks Data Scientist, Sanne, and Data Engineer, Don, began analyzing the best context converting and word order for ad texts – and the practicalities of a machine learning text generator.

The Power of Data-Driven Decisions

It was no small task for Sanne and Don to empower Social Blue to make data-driven decisions. Up until now, the lead generation expert had based their decisions primarily on experiences and intuition, which lead to incomplete evaluations and inaccurate result analyses. The Young Mavericks trainees guided Social Blue joyfully through their journey towards data-driven decision-making and campaigns. 

Sanne: “During this project, Don and I managed to combine our strengths and different expertise; we came up with many solutions together for Social Blue and we did the majority of the coding side-by-side.” Their biggest challenges were the time limitations – “We had just a month to develop a standalone tool” and communication with stakeholders. Don: “We have no shortage of technical expertise, but specific domain knowledge is an entirely different story. It was essential to ask the right questions in order to learn more about Social Blue’s needs, customers and markets.”

Working Together, Creating Together

Expectation management and a clear division of roles were also crucial. Don: “Of course, you want to show the far-reaching potential of data science, but ultimately the goal was to develop a concrete application for Social Blue.” To increase the chances of success, Sanne and Don always aligned their ideas, wishes and requirements in advance. Throughout the project, the Data Scientist and Data Engineer kept all Social Blue stakeholders in the loop with daily meetings, immediately sharing new functionalities and incorporating as much feedback as possible. Sanne: “The collaboration with Social Blue was wonderful. They were very enthusiastic, open to new ideas and interested in the data process.”

Developing a User-Friendly Application

The intensive, one-month project had Sanne and Don developing an application that every employee within Social Blue – with or without data expertise – can now use. Sanne: “By offering a clear explanation, we ensured that the use and outcome of each functionality was clear to all Social Blue employees.” As a result, the Young Mavericks’ application enables Social Blue to manage their own data-driven campaigns in the future. The Proof of Concept has provided the company insights into what makes ads successful – and eventually the ability to generate new, even more effective ad texts through machine learning.

Infinite Potential

After the project, Young Mavericks and Social Blue came to the joint conclusion that the Proof of Concept was successful and above all promising. Don: “I am proud that we were able to deliver measurable results and a usable data-driven Proof of Concept for Social Blue. We have shown the value of data science and a data-driven approach.” Sanne adds: “And all of this in just one month! The application has already been put into practice and the results are promising.”

In their final report, Social Blue indicated that their partnership with Young Mavericks has provided them with valuable insights into the endless possibilities of machine learning and how results from their previous campaigns are a major source of inspiration for creating successful new ads: ‘Looking back at the results, we see that there is still room for improvement, but it is evident that the machine learning tool has the potential to take the online advertising world to a completely new level.’

Want to collaborate with our data experts? Contact Young Mavericks

or learn more about our projects

ELT, ETL and Data Pipelines: loading Data in an Automated Manner and how to do it yourself

My name is Don and I am something between a data engineer and a data scientist. I automate repetitive tasks, generate insights from the data, manage projects and I often help in an advisory role for these processes. I especially enjoy the creative process required to solve problems using data.

Read more

Data Pipeline Implementation: how to do it yourself

These instructions build on what has been discussed in ELT, ETL and Data Pipelines. In that guide, we discussed the problems that arise in storing and using data for a company. In response to those problems, we introduced the concept of Data Pipelines, which helps the company become better aware of the data loading steps and incorporating these steps in the most optimal way to create a Data-driven Culture. We also discussed some specific tooling that can be used to properly deploy Data Pipelines. 

Now that we understand the concepts behind Data Pipelines, we will now apply them to implement a functioning Data Pipeline. Just like most of our data engineering processes, we follow a step-by-step plan and provide an implementation strategy for each step. 

Hopefully a step-by-step plan will give you a solid foundation when you are constructing your own Data Pipeline as well as the implementation methods.  You can find the whole code on our Giftlab.

Read more