Predicting sales funnel with a customer-relationship-management tool

Contents

1. BUSINESS CHALLENGE

2. TECHNICAL APPROACH

2.1. Information sources

2.1.1. Company information

2.1.2. Internal company data

2.2. Artificial Intelligence

2.2.1. Machine Learning techniques

2.2.2. Deep Learning techniques

2.3. Tested architectures

2.4. Final architecture

3. CONCLUSION

3.1. Objectives revision and functional requirements

3.2. Future guidelines

Research Objectives and Core Themes

The primary objective of this bachelor's thesis is to develop a predictive model that identifies sales opportunities within the B2B sector that have a high propensity to be lost, enabling sales personnel to intervene proactively. The research explores the integration of internal CRM data with external company information sourced via social network APIs to improve forecasting accuracy and support revenue retention.

• Development of a predictive customer retention model for the B2B telecommunications sector.
• Implementation of data enrichment strategies using public social network APIs to overcome internal data gaps.
• Empirical evaluation of various machine learning and deep learning architectures, including ensembles.
• Design of a Graphical User Interface (GUI) to streamline the application of predictive analytics in a business environment.

Auszug aus dem Buch

Summary of Chapters

1. BUSINESS CHALLENGE: Discusses the growing importance of Artificial Intelligence in business, the shift toward B2B forecasting, and the role of CRM systems in tracking sales opportunities.

2. TECHNICAL APPROACH: Details the sourcing of external and internal data, the exploratory analysis performed on variables, and the comparative study of Machine Learning and Deep Learning architectures.

3. CONCLUSION: Reviews the project's success in achieving high recall in predicting lost opportunities and outlines recommendations for future implementation and data usage.

Key Terms

Artificial Intelligence, Machine Learning, Deep Learning, Sales Funnel, Customer Retention, Churn Prevention, B2B, Predictive Analytics, Data Enrichment, LinkedIn API, Ensemble Models, Neural Networks, Feature Engineering, CRM, Forecasting.

Frequently Asked Questions

What is the core purpose of this research?

The study aims to create a predictive system for a telecommunications company to identify B2B sales opportunities at high risk of failure, allowing sales teams to react quickly and reduce churn.

Which sectors and companies are targeted by this model?

The research focuses on a large company in the telecommunications sector, specifically investigating the B2B sales funnel in the Argentinian market.

What is the primary research goal?

The goal is to maximize recall in order to capture the highest possible number of failing opportunities, thereby enabling effective preventive sales actions.

Which specific methodologies were applied?

The author employed a combination of traditional machine learning algorithms (e.g., Random Forest, Gradient Boosting) and deep learning techniques (e.g., MLP, LSTM) within an ensemble framework.

What is the main focus of the internal data analysis?

The analysis covers the sales funnel states and various contract-related variables, cleaning the data for noise and preparing it for predictive modeling.

How is the model evaluated?

The performance is evaluated primarily using recall and precision metrics, with a specific emphasis on the F1-score to balance the trade-off between identifying failures and avoiding false alarms.

How does the project handle data protection?

The project utilizes publicly available, GDPR-compliant data from LinkedIn to enrich the internal company information, ensuring that commercial use is legally permissible.

What role does the GUI play in the research?

The custom-built Python-based GUI simplifies the complex data gathering and predictive execution processes, making the developed tools accessible and user-friendly for business applications.