Identifying the profit margin on moving 130,000 loads of freight each month, spread across 600 trucks, was a difficult task for Queensland transport firm Followmont Transport.
Pricing pressure from larger competitors such as transport giant Toll Group, meant that Followmont needed to better understand how to maximise revenue from its freight movements. The company also needed to quickly identify why certain customers were taking their business elsewhere.
“Margins are always tight in transport,” says CIO Paul Smith. “Determining the margin on a load of freight is very difficult without knowing the consignment note details on that load, and it does vary depending on the size of the cartons and pallets.
“Followmont has been a successful business for 30-odd years; however, not knowing where we are missing out on revenue… was what we decided we needed to tackle.
“We didn’t understand where we were losing out,” he says. “There’s too much data; we carry 2 million consignments – how do you know which one is going to be losing you money?"
Followmont uses a custom-built transport management system - developed by Brisbane firm Omnix - to generate freight data but the software doesn't do data analysis.
Deploying a business intelligence system that would provide deep insights into company-wide data was the answer. In late 2013, the company deployed SAS Visual Analytics to enables executives and branch managers to analyse information on the fly using real-time data visualisations.
The initial roll out was not without its teething problems – achieving the best performance from a BI tool running in a virtual server environment was a challenge, says Smith.
“We were very much on the edge of where they [SAS] wanted us to be on the hardware. For example, traditionally SAS will sell you a blade infrastructure with a dedicated physical host.
“We’re [running] a fully virtualised environment like a lot of other SME organisations … we said to them ‘this needs to run in a virtualised environment, we can’t afford to have dedicated hardware,’” he says
SAS was able to tune the system to work with Followmont’s virtualised environment.
“It was the toughest part of the implementation from our point of view,” Smith says.
“When you’re selling a $100,000 product and then saying you need to run it on $120,000 [worth of] hardware, most IT managers and CIOs are going to baulk at that, and say ‘you need to work within our existing infrastructure.’
“Installing it was fine but getting the performance out of it was another story because there’s a lot of tuning [that needed to be] done with Linux, JBoss, and the actual VMware host and SAN etc. They [SAS] worked with us to their credit.”
Better service at a lower cost
Followmont’s sales staff use the analysis software on their iPads to gain quick access to historical customer data when they are on the road.
“It shows them 12 month history, revenue, kilos, consignment notes – they can look at all the freight … they can walk in with a decent knowledge of how the account is going,” he says.
“If they see it’s constantly on the increase and we must be doing a good job, they can go in there and give them some sharper rates to reward them for pushing extra business to us. If we see [sales] declining, we can say ‘look, are there any issues that we’ve got and how can we assist you further?’”
The software also helps Followmont keep a lid on costs by including wage data in its analysis when identifying, for instance, how much it costs per kilogram to deliver freight to certain areas.
“We can concentrate on areas [where costs are] a bit high so we can put a bit of pressure on our branch managers to manage their wages more effectively when we see the dollars creeping up per kilo.”
Keeping an eye on staff
The software is also used to analyse annual and sick leave trends among staff members to identify patterns around absenteeism and annual leave.
According to Smith, human resources is one area that lends itself to analytical tools particularly when information about employees is being correlated with data about wages, freight movements, and trucks.
“You’ve got to sit there and say ‘here’s all my wage, freight and fleet data on my trucks, how can we correlate this?’
“We can bring in some fuel usage data and cross-reference that with data around what we are delivering into Rockhampton, for example, and work out if usage per kilo of freight delivered is too high.
“We can determine if we need to shuffle our fleet around to give them smaller units if they are using large trucks that are basically empty to deliver freight,” he says.
Followmont is also convincing truck mechanics to look at their fleet history, kilometres driven and projected kilometres over a 12-month period. This helps the company identify when to rotate trucks.
“If you’ve got leased trucks you want them to have the same kilometres at the end of the lease. You don’t want to have one with 3 million [kilometres] and one with 300,000. So we can rotate fleet around depending on usage.”
Follow Byron Connolly on Twitter:@ByronConnolly