Succes story
Young Capital and Relevant Online utilise the power of AI. Driving recruitment efficiency: predictive matchmaking model.
YoungCapital is a recruitment agency with over 20,000 candidates at work every week. YoungCapital connects the new generation of employees with employers by managing the recruitment and selection process, as well as handling contracts and salary payments. It has the largest database of young people in Europe and its growth has been unstoppable in recent years.
The challenge
YoungCapital faces a major challenge in efficiently matching job seekers with employers through online advertising. Managing marketing budgets across different channels and campaigns is difficult and labor-intensive, with unpredictable demand for vacancies adding complexity.
Existing tools like Google’s “Performance Max” or other programmatic solutions focus only on budget optimization within their own platforms, lacking the ability to match candidates to employers’ needs effectively. To address this, YoungCapital and Relevant Online developed the Predictive Automated Matchmaking Model, aimed at finding the most cost-effective channels to match candidates with employers, minimizing the “cost per hire.”
“By combining YoungCapital’s knowledge of recruitment with Relevant Online’s AI expertise, the coöperation revolutionises the procurement of candidates, resulting in increased efficiency, a significant lower cost per hire and enhanced customer- and candidate satisfaction.”
Peter Segerius, Senior Online Marketer at YoungCapital
The solution
The model is trained on data from job openings, applications, and Google Analytics 4. While it is often assumed that more data is better, the optimal amount of data was identified from both business and machine learning perspectives. The model predicts the number of job applications for job openings each week and is retrained weekly with the latest data.
The dataset, built by Relevant Online in collaboration with YoungCapital, includes job openings, categories, locations, applications, and marketing data from Google Analytics 4. This enables tracking candidates through YoungCapital’s system, from the initial acquisition channel to the final contract signing. Only the most relevant features of the dataset are used to improve prediction accuracy. This “feature extraction” ensures greater transparency and explainability compared to “black-box” neural networks, with traceability being a key focus.
The data is stored in a custom-built data warehouse designed and developed by Relevant Online over the years. The project leverages Google Cloud Platform (GCP) infrastructure, including BigQuery for data storage, Google Cloud Storage (GCS) for pipeline metadata, and Dataproc for preprocessing and training the model. Maintenance involves testing, adding features, and monitoring the pipeline for consistent performance.
Impact
The model goes beyond candidate acquisition by identifying opportunities in the market, such as scarcity of candidate profiles, job availability, difficulty in finding suitable profiles, and associated revenue. It provides value across multiple departments. For the marketing team, it improves the allocation of ad spend across channels. Sales and account teams use its insights on candidate scarcity and pricing to inform customer negotiations. Recruitment benefits from the ability to deliver the right candidates at the right time, and the finance team gains a clearer understanding of marketing cost drivers and financial outcomes. The model optimizes ad spend and cost-per-hire, providing actionable insights across multiple channels.
The model delivers accurate predictions with full transparency in how data influences outcomes. Continuous optimization ensures long-term benefits, including enhanced cost control and improved marketing decisions.
Succes story
Young Capital and Relevant Online utilise the power of AI. Driving recruitment efficiency: predictive matchmaking model.
YoungCapital is a recruitment agency with over 20,000 candidates at work every week. YoungCapital connects the new generation of employees with employers by managing the recruitment and selection process, as well as handling contracts and salary payments. It has the largest database of young people in Europe and its growth has been unstoppable in recent years.
The challenge
YoungCapital faces a major challenge in efficiently matching job seekers with employers through online advertising. Managing marketing budgets across different channels and campaigns is difficult and labor-intensive, with unpredictable demand for vacancies adding complexity.
Existing tools like Google’s “Performance Max” or other programmatic solutions focus only on budget optimization within their own platforms, lacking the ability to match candidates to employers’ needs effectively. To address this, YoungCapital and Relevant Online developed the Predictive Automated Matchmaking Model, aimed at finding the most cost-effective channels to match candidates with employers, minimizing the “cost per hire.”
“By combining YoungCapital’s knowledge of recruitment with Relevant Online’s AI expertise, the coöperation revolutionises the procurement of candidates, resulting in increased efficiency, a significant lower cost per hire and enhanced customer- and candidate satisfaction.”
Peter Segerius, Senior Online Marketer at YoungCapital
The solution
The model is trained on data from job openings, applications, and Google Analytics 4. While it is often assumed that more data is better, the optimal amount of data was identified from both business and machine learning perspectives. The model predicts the number of job applications for job openings each week and is retrained weekly with the latest data.
The dataset, built by Relevant Online in collaboration with YoungCapital, includes job openings, categories, locations, applications, and marketing data from Google Analytics 4. This enables tracking candidates through YoungCapital’s system, from the initial acquisition channel to the final contract signing. Only the most relevant features of the dataset are used to improve prediction accuracy. This “feature extraction” ensures greater transparency and explainability compared to “black-box” neural networks, with traceability being a key focus.
The data is stored in a custom-built data warehouse designed and developed by Relevant Online over the years. The project leverages Google Cloud Platform (GCP) infrastructure, including BigQuery for data storage, Google Cloud Storage (GCS) for pipeline metadata, and Dataproc for preprocessing and training the model. Maintenance involves testing, adding features, and monitoring the pipeline for consistent performance.
Impact
The model goes beyond candidate acquisition by identifying opportunities in the market, such as scarcity of candidate profiles, job availability, difficulty in finding suitable profiles, and associated revenue. It provides value across multiple departments. For the marketing team, it improves the allocation of ad spend across channels. Sales and account teams use its insights on candidate scarcity and pricing to inform customer negotiations. Recruitment benefits from the ability to deliver the right candidates at the right time, and the finance team gains a clearer understanding of marketing cost drivers and financial outcomes. The model optimizes ad spend and cost-per-hire, providing actionable insights across multiple channels.
The model delivers accurate predictions with full transparency in how data influences outcomes. Continuous optimization ensures long-term benefits, including enhanced cost control and improved marketing decisions.
Succes story
Young Capital and Relevant Online utilise the power of AI. Driving recruitment efficiency: predictive matchmaking model.
YoungCapital is a recruitment agency with over 20,000 candidates at work every week. YoungCapital connects the new generation of employees with employers by managing the recruitment and selection process, as well as handling contracts and salary payments. It has the largest database of young people in Europe and its growth has been unstoppable in recent years.
The challenge
YoungCapital faces a major challenge in efficiently matching job seekers with employers through online advertising. Managing marketing budgets across different channels and campaigns is difficult and labor-intensive, with unpredictable demand for vacancies adding complexity.
Existing tools like Google’s “Performance Max” or other programmatic solutions focus only on budget optimization within their own platforms, lacking the ability to match candidates to employers’ needs effectively. To address this, YoungCapital and Relevant Online developed the Predictive Automated Matchmaking Model, aimed at finding the most cost-effective channels to match candidates with employers, minimizing the “cost per hire.”
“By combining YoungCapital’s knowledge of recruitment with Relevant Online’s AI expertise, the coöperation revolutionises the procurement of candidates, resulting in increased efficiency, a significant lower cost per hire and enhanced customer- and candidate satisfaction.”
Peter Segerius, Senior Online Marketer at YoungCapital
The solution
The model is trained on data from job openings, applications, and Google Analytics 4. While it is often assumed that more data is better, the optimal amount of data was identified from both business and machine learning perspectives. The model predicts the number of job applications for job openings each week and is retrained weekly with the latest data.
The dataset, built by Relevant Online in collaboration with YoungCapital, includes job openings, categories, locations, applications, and marketing data from Google Analytics 4. This enables tracking candidates through YoungCapital’s system, from the initial acquisition channel to the final contract signing. Only the most relevant features of the dataset are used to improve prediction accuracy. This “feature extraction” ensures greater transparency and explainability compared to “black-box” neural networks, with traceability being a key focus.
The data is stored in a custom-built data warehouse designed and developed by Relevant Online over the years. The project leverages Google Cloud Platform (GCP) infrastructure, including BigQuery for data storage, Google Cloud Storage (GCS) for pipeline metadata, and Dataproc for preprocessing and training the model. Maintenance involves testing, adding features, and monitoring the pipeline for consistent performance.
Impact
The model goes beyond candidate acquisition by identifying opportunities in the market, such as scarcity of candidate profiles, job availability, difficulty in finding suitable profiles, and associated revenue. It provides value across multiple departments. For the marketing team, it improves the allocation of ad spend across channels. Sales and account teams use its insights on candidate scarcity and pricing to inform customer negotiations. Recruitment benefits from the ability to deliver the right candidates at the right time, and the finance team gains a clearer understanding of marketing cost drivers and financial outcomes. The model optimizes ad spend and cost-per-hire, providing actionable insights across multiple channels.
The model delivers accurate predictions with full transparency in how data influences outcomes. Continuous optimization ensures long-term benefits, including enhanced cost control and improved marketing decisions.
