Optimizing Reconditioning Costs in Used Automobiles: A Data Science Approach
Company Information
Canada Drives aims to simplify the process of buying and selling cars in Canada, positioning itself as the most convenient option for consumers in the automotive market. This mission reflects a commitment to leveraging technology and customer service to make the car buying and selling experience as straightforward and hassle-free as possible. By offering a streamlined online platform, Canada Drives enables customers to quickly find, buy, or sell their cars without the traditional complexities and time-consuming aspects of automotive transactions. This approach not only caters to the evolving expectations of consumers for online services but also aims to transform the traditional car dealership model, making it more accessible and user-friendly for Canadians across the country.
The Challenge
Imagine a bustling used car market, vibrant and dynamic, but riddled with the complex challenge of accurately assessing reconditioning costs for various makes and models. This was the reality for Canada Drives, a pivotal player in this marketplace. Enter ProCogia, a beacon of data science innovation, tasked with transforming the landscape of cost allocation through sophisticated algorithmic redesign. This story unfolds as a testament to the power of data science in reshaping industry standards.
Canada Drives faced a significant hurdle: the allocation of reconditioning costs across a diverse range of used automobiles was imprecise, leading to inconsistent pricing and efficiency issues. This inaccuracy impacted their business operations, demanding an urgent and effective solution.
The Results
ProCogia’s data science team, leveraging their expertise in Advisory, Data Science, BI & Analytics, and Data Engineering, using tools like Python, SQL, and AWS, approached this challenge with a new idea. They recommended integrating mechanic notes as a novel data layer, enhancing the training of their models. This approach not only provided a more granular understanding of each vehicle’s condition but also introduced a previously untapped data source: car history and mechanical information.
The redesigned recon models incorporated this new data set, transforming qualitative mechanic insights into quantifiable metrics through a sophisticated vector matrix. This enhancement allowed for a more nuanced and accurate prediction of reconditioning costs. The result was a model that performed 50% better than its predecessor, with a D2 Tweedy score improvement from 0.26 to 0
The enhanced recon model had a tangible impact on Canada Drives’ operations. It not only streamlined the cost allocation process but also introduced a level of precision that was previously unattainable. This improvement translated into better pricing strategies, increased operational efficiency, and a competitive edge in the used car market. In an innovative twist, ProCogia demonstrated the potential of integrating computer vision to assess the condition of used vehicles. This technology could revolutionize various aspects of the buying and selling journey, offering insights that are both rapid and reliable.
Procogia’s Approach
Understood the data for training the recon model – from source to feature engineering to the final state.
Studied the incumbent Recon model to get a better understanding of what improvements could be made.
Recommended the client use mechanic notes as an additional layer for training a more accurate model. Our approach was to introduce more data sources to the recon model. One data set the client had not previously considered was a car history/mechanical information history.
Built an enhanced recon model that used pre-existing training data, which trains the algorithm to predict outcomes in the design/model. At the same time, integrating a vector matrix to process qualitative data into quantitative data to build and enhance the model.
The Detail
This architectural redevelopment has had positive effects across the organization, from optimizing where the products are sold, to how sizes are displayed on the client’s e-commerce site. Improved capabilities include:
The enhanced model performed 50% better than the base model. D2 Tweedy score of 0.4 as compared to 0.26 for the base model.
Delivered a set of strategic questions – that used the important words and features from the improved model – to ask the seller more pertinent questions that would help better estimate the price of the vehicle being sold.
We recommended a new set of strategic questions for sellers to answer on their website to capture their customers’ needs and preferences more accurately, as well as adapt to seasonal trends.
We demonstrated through computer vision how you can assess the condition of used vehicles, which could be integrated into different components of the buying/selling journey.
Conclusion
The collaboration between Canada Drives and ProCogia underscores the transformative potential of data science in the automotive industry. By harnessing the power of advanced algorithms and diverse data sets, companies can not only solve complex challenges but also drive innovation and efficiency. As we witness the evolution of data science, one wonders: what other industries could benefit from such a profound technological integration?
Explore more stories
Dig deeper into data development by browsing our blogs…
Implemented Bookmarking Apps
A leading investment consulting firm faced challenges in managing and sharing complex data analyses through their internal tools developed with Shiny. As the firm sought
Optimizing and Scaling RESTful API for an Investment Consulting Firm
Optimizing and Scaling RESTful API for an Investment Consulting Firm Company Information A prominent investment consulting firm sought to elevate the performance and maintainability
Low-code solutions using Azure & Snowflake
ProCogia implemented low-code solutions using Azure Data Factory and Snowflake for the Performance Data Science team of a Leading Sports Organization
Get in Touch
Let us leverage your data so that you can make smarter decisions. Talk to our team of data experts today or fill in this form and we’ll be in touch.
