Is Home Delivery the Long Way Around?

This article first appeared in the Winter 2017 edition of Motion Magazine.

It’s rare to drive down a residential street without seeing at least one vegetable liveried vehicle. But is grocery home delivery causing more congestion? We decided to look into it.

Our consumer habits have changed a great deal. Even over the last few years the number of us shopping for groceries online has surged.

We wanted to figure out what impact this might be having on road congestion, greenhouse gas emissions and shoppers’ time. Are more home deliveries leading to a greater cumulative total number of miles driven for our groceries – or would we be better off each travelling to the store to fill up our own cars?

Our intrigue got the better of us. We decided to crunch some anonymous numbers and look at the data we have for one of our supermarket customers. Our aim was to identify whether, on balance, home deliveries seem to be a force for good or evil on our roads.

At no point would we argue that this is a scientifically robust review. It’s more a sideways glance at data from one specific day, in a handful of locations, up and down the UK.

Growing Online Demand

Here in the UK we are among the savviest in the world when it comes to internet retailing. Research carried out by OC & C Strategy Consultants has identified the seismic shift currently underway.

In 2005 91% (£152bn) of sales in the UK were made in store. By 2015, this figure had dropped to 78% with almost a fifth (£35.9bn) of sales being made online for home delivery. The research predicts that by 2025 home delivery could be responsible for 30% of sales and click-and-collect for 10%, whilst in-store purchases will decline to 60%.

These figures accepted, it is important to differentiate grocery home deliveries from other online shopping. Food deliveries are inherently more efficient because they are batched and scheduled. Other online shopping is more random because the online buying is more unpredictable as it doesn’t have the same level of planned delivery windows as with food.

That said, the grocery sector has arguably seen the most rapid growth, with research released earlier this year from Mintel showing that 48% of Britons are already online grocery shoppers. So the UK is decidedly invested in the home delivery evolution – is that good or bad for our roads and environment?

The Data

We work with several of the top supermarkets, helping them to manage their home delivery operations, making them as efficient as they can be, and ultimately delivering the very best customer service.

For our sideways glance, we looked at more than 12,000 miles worth of data and multiple routes in Aberdeen, Inverness, Leeds, West Lothian, Nottingham, Romford and St Austell. We wanted to get a snapshot from North to South and from the urban to the rural. For each drop we used a Google API to calculate the distance to the nearest supermarket of the same brand.

This way we could compare how far a person might need to travel in order to visit the store themselves with how far the home delivery vehicle actually travelled in reality. The API used data from the supermarket’s website to locate the nearest full sized shop.

What we Found

Looking at the total figures, shoppers would have travelled 33% further, if each individual had travelled to their nearest supermarket themselves.

The circa 12,000 mile round trips by home delivery vehicles would have equated to nearly 19,000 miles had each individual made their own journey.

Of the 221 home delivery journeys we looked at, we estimate just 47 (21%) of the vehicles would have travelled further than if the customer made their own outing. Sometimes this was by margins of just 1 or 2 miles – at other times by tens of miles.

Of the sample we investigated, the average number of drops per route for the home delivery drivers was 14 – though this did vary depending on the location. For example, in Leeds, the average number of drops was 16, whilst in St Austell in Cornwall, the average was 10.

Extrapolate and Listen

I am the first to acknowledge the limitations of this flirtation with data analysis. All of the data we used was for a single day, looking at just a handful of distribution centres.

When calculating the nearest store in order to make a comparison with our tracked home delivery data, we relied on Google to tell us the distance to get there and back. We’re sure this data is more or less accurate – but couldn’t check through each instance for validity. We also know that the story is more complicated than this – with people often visiting the supermarket as part of another journey.

Or maybe they would have visited a different brand of supermarket if they weren’t shopping online. All that said, yes, we are making assumptions about what the consumer’s trip might have been, but we know absolutely what the home delivery vehicles did. And despite all of these drawbacks, we think this is still food for thought.

A whole third less mileage might have been travelled, showing the potential that Home Delivery has to reduce the burden on the road network and cut wasted mileage by sharing resources. Just think of all of the time saved by shoppers too. Taken overall, and with the assumptions mentioned, 79% of the grocery deliveries would be more efficient considering miles travelled by the route of the van versus the individual journeys taken by shoppers.

One delivery vehicle shared by 14 customers on average – rather than one car each making its own journey must logically be more efficient. You would think this would result in less congestion on the roads. And that’s a good job – given how invested we are, as a country, in this new form of retail therapy.