How we supported a freight transportation provider to optimize their pricing process by reducing the time and effort needed to calculate the price for each quote.
OPTIMISED
PRICING PROCESS
BOOSTING DEAL VELOCITY
& TEAM EFFICIENCY
ACHIEVING BUSINESS TRANSFORMATION
THE SITUATION
Our client, a freight transportation provider, needed support in driving business transformation towards data-driven decision making. Their main priority was automating and optimizing the pricing of orders, which at that time was done in a time-consuming and inefficient way. The method of pricing used by Transport managers was also very internal focused and without a data driven way to look at market indicators.
The goal was to reduce the time to quote, along with the effort required by transport managers to calculate the price.
THE SOLUTION
Amplify created and deployed a machine learning model to enable intelligent pricing of orders based on key factors:
- Load and unload country
- Route distance
- Time of year (additional features were derived to capture seasonality)
- Cost (considering fluctuation of fuel prices)
- Transport Platform supply and demand indicator
The final solution is a user-friendly app, where the transport manager will have to input the parameters featured in the model and instantly obtain a predicted price.
THE METHODS USED
We performed an extensive data analysis on all the different routes in order to identify major trends and disruptions in the data.
The freight transportation industry has seasonal trends but is also impacted by other events such as Covid-19 or Brexit.
After the data exploration phase a machine learning model was developed to predict the price. We opted for a tree-based gradient-boosting algorithm which has the advantage of faster training speed and better accuracy.
MACHINE LEARNING MODEL DEPLOYMENT
We deployed the machine learning model in a user-friendly app with an added authentication feature.
The transport managers are able to enter basic information of the load and unload zip and are able to get a recommended price (keeping in mind also Industry factors, route, empty kms).
They are also able to view the market price and compare to their own prices for a given route.

Technical Solution: Machine learning model deployed in an app to be used by the transport managers
BUSINESS IMPACT
By reducing the time and effort needed to calculate the price for each quote, transport managers were able to maximize the full value of their orders and speed them down the sales funnel. This in turn extended their capacity to take on a greater volume of orders.
More than this, price optimization helped our client achieve business transformation and enabled them to execute data-driven decision-making.