The Marginal Gains Revolution in Final Mile Logistics

by Sales, Innovation & Research Director David Cockrell


bikesMarginal Gains have been in the spotlight for a while, but in recent years the concept has been elevated to new heights within sport, with notable exponents being British Cycling.

Sir David Brailsford referred to the concept of the “aggregation of marginal gains” a number of years ago which was highlighted in some detail in a documentary about Team Sky, who stated that “you can achieve optimal performance by the aggregation of marginal gains. It means finding a 1% margin for improvement in everything we do”.

But where can you achieve ‘Marginal Gains’?



1. Address cleansing

If you’re running a multi-million pound home delivery operation employing hundreds if not thousands of drivers, leasing their vehicles, paying fuel, insurance and maintenance costs, then surely you owe it to yourself to get the little things upstream correct, as they can cause huge impacts further downstream in the execution phase.

  • Ensure that your customer addresses are 100% correct
  • Confirm that your addresses and postcodes match


2. Premise-level geocoding

This eradicates the excuse for delivery failure through not finding the address and the mythical ‘you weren’t in’ calling card being dropped through the letterbox of a fully staffed office of 60 people with a full time receptionist.

  • Using premise-level geographic co-ordinates for all delivery locations, your company can safely save around 30 seconds per delivery.
  • This level of accuracy is most felt when sending these co-ordinates to the satnavs and MDTs in your vehicles, meaning your drivers won’t have to hunt for addresses long before they come up.
  • This degree of clarity boosts job satisfaction and makes for happier drivers.
  • If using a courier with a mobile device, you can route them on foot to the exact building.


3. Using premise-level co-ordinates with satnav and in-cab mobile data terminals

Once you have appended premise-level coordinates to your customer address, whether this is a static or churning database, you will need to use it widely. Don’t just use it in route optimisation software, also send it through as part of the delivery manifest and sequence to satnav devices and mobile data terminals (MDT).

A first step would be to find out if you can send route manifests over the air to the device within your business or if you are still reliant on the manual adding of postcodes by the driver.


4. Location intelligence/location analytics

Knowing that your delivery is to a detached, semi-detached, terraced or self-contained flat is invaluable for a number of very practical reasons:

  • Allocate a two-man crew if certain heavy deliveries are to a self-contained flat in a block of flats
  • It’s easier to park outside a detached or semi-detached house than a terraced street or block of flats
  • The customer is on a Red Route, meaning that your drivers cannot park outside, and consequently will need to park around the corner and spend more time on foot
  • We can even tell if one side of a semidetached property has been converted into flats or several houses in terraced streets are now flats
  • Know beforehand if any of your delivery addresses are above commercial locations


5. Street-level mapping (Open Source, Closed Source and Cloud)

When it comes to the mapping used within a scheduling system, take the time to fully understand what has been deployed as standard – the newness of the maps, their coverage area and to what level of detail.

Open Source data including mapping is now a ubiquitous fixture in the business landscape, and using a FREE data source can seem a jolly good idea when faced with budgetary cuts or pressure to get more from your allocated budget, but FREE at source may not equate to FREE use in your business.

How truly routable is the data? Do you, for example, need to create new fields to work with your optimisation system? What guarantees do you have if it ‘goes wrong’, and who ensures version management? When licensing from a supplier, all of this is taken care of and you are supported technically, allowing for great overall peace of mind.

At Mapmechanics, we ourselves use Open Source data. We do however apply sophisticated routines to fix errors, create new fields and make these maps fully routable for various software systems in a range of formats and structures.

Closed Source mapping is still the most commonly used, with its ISO standards, large R&D budgets and rigorous testing, and that’s all before Mapmechanics undertakes further enhancements and fixes to make it the very best it can be.

Access to Cloud mapping as software as a Service (SaaS) via APIs is a growing requirement with businesses looking to outsource time and distance matrices to cloud services utilising road speed data by time of day, day of week, live traffic, truck width, weight and height attributes as well as HAZMAT requirements.

This is backed up with the ability to undertake live ETAs (estimated time of arrivals) based on current vehicle locations and destination location with a Live Traffic feed.

Currently cloud solutions only allow for a limited range of scenarios of destinations and depots, and are therefore not yet suitable for large logistics operations which are the preserve of on premise cloud-hosted solutions of full VRS systems.


6. Enhanced Road Speeds

  • By time of day
  • By day of week
  • By direction of travel (bidirectional)


7. What’s to come

I believe the next big battleground for logistics routing software providers will be the individual routes between drops. Depending on what vehicle routing and scheduling system you are using, edits and modifications to the underlying maps used can force the routing engine into behaviour changes not overtly possible through the software front end.

If you want to get the strategic and operational edge over your competitors, reduce inefficiencies and increase logistics fleet efficiencies on a shift by shift, day by day basis, these are all excellent starting places.

I am already working with a number of businesses to address all the points raised above. If this is not on your own agenda, I would strongly encourage you to add them quickly.

Contact me via LinkedIn, Twitter @PureGeoSpatial or email to discuss any of the above points in more detail. We also offer ‘Play with Data’ sessions on Friday mornings to explore any of these topics in a more ‘hands on’ way.