Publication
Optimizing Sales Performance in Creative as a Service (CaaS) Companies: A Machine Learning Approach to Opportunity Time-Series Forecasting
| Summary: | This thesis addresses the gap in empirical studies on the sales generation process of Creative as a Service (CaaS) companies across various channels. The study employs predictive analysis techniques, including feature selection and hyperparameter optimization, to develop a supervised machine learning model tailored to untraditional sales approaches that incorporate unique parameters and time-series data. The objective is to identify critical attributes within the sales funnel that significantly impact the forecasting process in the B2B CaaS business model. The paper explores different types of machine learning, with a particular focus on supervised learning, and its applicability in decision-making and business forecasting. To address the objective, a customized model-stacking approach is proposed, leveraging ensemble methods such as boosting, trees, random forests, and neural networks. The proposed methodology is presented within the framework of CRISP-DM, encompassing phases such as data preparation, cleaning, transformation, and modeling. The challenges associated with missing and categorical data, as well as the importance of feature selection and encoding in B2B sales forecasting, are also examined. Methodologies for opportunity forecasting in CaaS sales, the significance of sales pipeline management, and the utilization of time series forecasting are discussed. Building upon previous research on sales win probability, this study introduces a novel approach to forecasting win probability in B2B sales. By identifying critical attributes within the sales funnel that significantly influence the forecasting process, valuable insights are gained for improving the sales funnel and process through a comprehensive analysis of the CaaS company lifecycle. The findings contribute to a deeper understanding of the dynamics involved in generating sales for CaaS companies and offer practical implications for enhancing the effectiveness of sales strategies and decision-making processes. |
|---|---|
| Main Authors: | San, Suha |
| Subject: | B2B Sales Predictive Modeling B2B Sales Funnel Machine Learning Creative as a Service (CaaS) Sales Forecasting SDG 8 - Decent work and economic growth SDG 9 - Industry, innovation and infrastructure |
| Year: | 2023 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | open access |
| Associated institution: | Universidade Nova de Lisboa |
| Language: | English |
| Origin: | Repositório Institucional da UNL |
| Summary: | This thesis addresses the gap in empirical studies on the sales generation process of Creative as a Service (CaaS) companies across various channels. The study employs predictive analysis techniques, including feature selection and hyperparameter optimization, to develop a supervised machine learning model tailored to untraditional sales approaches that incorporate unique parameters and time-series data. The objective is to identify critical attributes within the sales funnel that significantly impact the forecasting process in the B2B CaaS business model. The paper explores different types of machine learning, with a particular focus on supervised learning, and its applicability in decision-making and business forecasting. To address the objective, a customized model-stacking approach is proposed, leveraging ensemble methods such as boosting, trees, random forests, and neural networks. The proposed methodology is presented within the framework of CRISP-DM, encompassing phases such as data preparation, cleaning, transformation, and modeling. The challenges associated with missing and categorical data, as well as the importance of feature selection and encoding in B2B sales forecasting, are also examined. Methodologies for opportunity forecasting in CaaS sales, the significance of sales pipeline management, and the utilization of time series forecasting are discussed. Building upon previous research on sales win probability, this study introduces a novel approach to forecasting win probability in B2B sales. By identifying critical attributes within the sales funnel that significantly influence the forecasting process, valuable insights are gained for improving the sales funnel and process through a comprehensive analysis of the CaaS company lifecycle. The findings contribute to a deeper understanding of the dynamics involved in generating sales for CaaS companies and offer practical implications for enhancing the effectiveness of sales strategies and decision-making processes. |
|---|