Background and Challenges

A leading website domain wholesaler, which manages a portfolio of hundreds of domains and partners with 200 resellers such as GoDaddy and Namecheap, faced significant challenges in pricing and promotion management. The company had been manually making thousands of distinct pricing and promotional decisions, a time-consuming process that often yielded suboptimal results. Furthermore, the client’s extensive history of promotional campaigns had resulted in a wealth of data suitable for training machine learning models.

Approach

The client engaged L.E.K. Consulting to develop a sophisticated pricing engine that could dynamically recommend the optimal price and promotional strategies for each domain-reseller combination, aiming to maximize the lifetime value (LTV) of each pairing. The approach comprised three main analytical phases:

  1. Demand forecasting. We developed a Bayesian time series model to predict future demand for each domain-reseller combination over the next 12 months, considering seasonality and various market dynamics.
  2. Renewal propensity modeling. We constructed a light gradient boosting model using 27 key variables to forecast the likelihood of domain renewal. We refined this model to achieve predictions within approximately 1% accuracy for the most popular domains.
  3. Pricing optimization. We developed a pricing optimization model that integrated demand forecasts and renewal predictions, evaluating thousands of price and promotion combinations to maximize LTV. By adjusting prices, the model evaluated the impacts on sales and renewals against a baseline to identify the most-profitable strategies. It also incorporated an iterative A/B testing approach, enabling continuous refinement of recommendations as new data became available.

Results

The implementation of the pricing optimization engine led to significant improvements in operational efficiency and revenue.

  • Enhanced decision-making: The automation of pricing decisions freed up significant resources and reduced the likelihood of human error.
  • Increased revenue: The optimization strategies proposed by the engine were projected to increase annual revenue by approximately $20 million through improved LTV.
  • Accurate predictions: The demand forecasting model successfully predicted demand for over 90% of the domain portfolio within a 20% margin of the actual figures, while the renewal model predicted renewal rates within 1% of actual rates for high-volume domains.
  • Strategic impact: The detailed analytics and predictive capabilities of the engine allowed the client to not only optimize current offerings but also to plan strategically for future market conditions.

L.E.K. Consulting is a registered trademark of L.E.K. Consulting LLC. All other products and brands mentioned in this document are properties of their respective owners. © 2024 L.E.K. Consulting LLC

Cutting Costs by 40% in European Plasma Centers Through Data-Driven Strategies
doctors in meeting
Learn how L.E.K. Consulting synthesized insights from rigorous analyses into a four-phase process to optimize a leading blood plasma donation network's performance.