Home » Insights » Predictive analytics: Predictive modeling to control travel spend

Predictive analytics: Predictive modeling to control travel spend

|

Results at a glance

  • Data modeling revealed a probable cost increase valued at US $300,000 at company’s top supplier
  • Risk identified in key market (London), representing more than US $1.5 million spend
  • US $150,000 potential savings by moving bookings to 2 other suppliers

Challenge

A leading global technology company with a key workforce based in Western Europe wanted to better react to market forces that impact their supplier costs. We applied advanced modeling to their integrated travel, expense, credit card and HR data and combined it with CWT’s vast industry-specific data.

Our capabilities combine clients’ historic travel data sets with more than 85 macroeconomic KPIs, plus commodity prices such as GDP, inflation, food prices, and fuel prices. We then analyze this data to identify patterns and correlations that uncover robust predictions about different variables, including factors such as ‘number of trips’ and ‘cost per trip’.

The client wanted us to apply modeling to their activities and spend in London because it is a key market with a high volume of business travel. London represents spend of more than US $1.5 million each quarter for this client.

Results

After identifying the probable risks for our client, we recommended they take actions to protect the organization from the impacts of a price increases at its top hotel supplier.

We recommended targeted communications to employees traveling to London, as well as modifications to the online booking tool to redirect travelers to the two alternate hotel suppliers with lower rates.

These actions would help the client reduce the number of bookings at its top supplier and the underlying cost increase. We helped the client save around US $150,000 of additional costs by shifting a significant portion of bookings to two alternative suppliers with lower costs over the same time period.