Big Data & Logistics (Part I): Breaking Down Big Data

The term ‘Big Data’ has always served to confuse with its broad ambit and unclear usage. At its core, Big Data is the utilisation of a tremendous amount of data sets, over a great variety of data types, to reveal relevant patterns and trends. The usage of data must be seen as a viable and necessary competitive advantage that companies can tap on, which ultimately translates into actionable implementations. In its raw form, corporates struggle to utilise such a huge amount of data, due to its lack of structure. Data collection by itself serves no purpose, unless it can be utilised with a practical perspective. Notwithstanding this, its adoption can enable companies to have a more complete picture of customer satisfaction and operations.

The booming data analytics sector in Singapore is expected to contribute at least 1 billion to the economy per year by 2017. In this regard, supply chains today are ever-growing, becoming more complex and sophisticated. With the onset of new technologies adopted in the logistics sector, massive amounts of data are being produced every day, in terms of volume, variety and velocity. The responsibility is on logistics companies to utilise this ever-increasing bundle of data streams to develop a unique competitive differentiation avenue. With the ‘last mile’ in logistics traditionally seen as the slowest & the least cost-effective part of their operations, the usage of Big Data to optimise last-mile deliveries is particularly attractive to third-party logistics providers and transporters.

The logistics industry is one that is prime for Big Data disruption. This can be attributed to the following reasons. Firstly, ‘last-mile’ logistics companies make multiple delivery trips on a daily basis. Such trips necessarily involves customer interaction. These characteristics allow logistic firms to create a data bank with regards to customer satisfaction. Secondly, the local presence of a fleet of transport vehicles is necessary for any third-party logistics provider to operate. This allows for automatic collection of local information on transport routes, which can be valuable in terms of generating demographic and traffic statistics. In addition, the underlying nature of the transportation of physical goods generates unavoidable inefficiencies that is ripe for Big Data to disrupt.

Broadly, value creation enjoyed by logistics companies stemming from the implementation of Big Data can be divided into 2 different categories – that of (1) Operational Efficiency and (2) Customer Satisfaction.

Operational Efficiency

Firms can hope to increase the level of operational efficiency with the analysis of large volumes of data. Third-party logistics providers who intend to become prospective users of Big Data will find the following potential grounds of application helpful.

Increasing Last Mile Efficiency

The essence of Big Data’s involvement in route optimisation is to optimise each driver’s daily route. This can be done by analysing the viability of delivery routes based on time windows, weather conditions, as well as considering the availability and locational data of recipients based on historical data. One such example to use data to improve route optimisation would be to compare the planned delivery time versus the actual delivery time to examine the deviation between the two. The aim would then be to minimise the deviation to ensure more accurate and efficient planning. This allows minimisation of unsuccessful delivery attempts for the next logistics run.


The performance of third-party logistic providers is commonly measured against various Key Performance Indicators (KPIs), such as DIFOT (Delivered In-Full, On-Time) and Order Accuracy. Incentive or negative structures set in place would incentivize transporters to meet KPIs regularly. Good management of data would help a company to fulfill their KPIs.
In addition, the consistent and thorough use of data analytics can help create industrial benchmarks for the performance of every third-party logistics provider. Each logistics provider would be able to analyse their own performance against others that are similar to them in the industry. Each logistics provider would then know where he stands in the terms of industry standards and can take necessary steps to improve the company. An industrial benchmark would be of value as it allows transparency as to the performance of each provider and serves as a bargaining chip for third-party logistics providers to win big contracts.

Customer Satisfaction

Businesses are able to harness the existence of a data bank to draw conclusions and actionable steps from a large & anonymous customer base. Big Data analytics enables companies to specifically target gaps in the consumer base, with the end goal to create targeted customer value.

Customer Loyalty – Maximising customer retention

The analysis of data sets could possibly allow the logistics firm to analyse the ‘Attrition Potential’ (How likely is the customer to switch the services from one logisitics firm to another) of its customers. This can be done by analysing data harnessed from relevant customer touch points (delivery to the end recipient, customer service inquiries). The fact that a customer has continuously experienced delayed shipments could ring alarm bells to the logistics firm to refresh customer relations. Such analysis should run across the entire customer base, allowing the provider to initiate proactive counter-measures and implement customer loyalty programs accordingly. This kind of preventive measures to prevent the souring of relationships would not be possible without having some kind of data set in which to measure current performance against.

On the flip side, when struck by peak periods or economic downturns, using data at hand, the logistics provider can manage customer expectations better. Analysing previous delivery trends and performance can help companies forewarn their customers, thereby maintaining good relationships with them, while at the same time prepare for the peak periods of downturns by allocating the appropriate resources to get through such periods.

Understanding transactional data

Companies can have a clearer picture of its delivery process by gathering appropriate transactional data, improving consumer satisfaction as well as operational efficiency. Firstly, assessment of consumer behaviour via data could pinpoint customers who are not at home during delivery hours. Companies can use such data to appropriately channel their assigned trips accordingly to minimise chances of ‘failed deliveries’. An example would be the delivery of groceries. Most people would want to be at home when they receive their groceries, especially when they have ordered wet goods. However, the people who usually order groceries online are white collar workers with a 9 am to 5 pm job. They are only home to receive their groceries at night, after 6 pm or on weekends. As a result of this study of consumer behaviour, many e-commerce players doing grocery deliveries have opened up delivery time slots at night after 6 or on the weekends instead of relying on the usual weekdays time morning and afternoon slots.

Secondly, through the in-depth analysis of (1) Driver Efficiency, (2) Current Traffic Conditions, (3) Consumer Behaviour, (4) Vehicle Health; companies can determine an accurate time-window for deliveries to occur. In addition, delivery site demographics can now be analysed to improve delivery experience. For example, delivering in densely populated parts of the city or to a high rise apartment would necessarily be a factor in considerations such as the need to climb stairs and difficulties in parking. With such data, companies could consolidate all jobs in such areas and assign these jobs to a more skilled driver who can climb all these stairs and navigate complex city parking. Hence, finding the right metrics is important to bring about a holistic analysis of the delivery process.

Understanding Customer Feedback

In an age where technological advancements are at the forefront, Big Data’s relevance in understanding and deciphering customer feedback should not be understated. There are now a plethora of avenues for people to voice their customer experience over a particular product – discussion forums, social networks etc. The challenge would be for companies to extract relevant feedback from a mountain of natural language content created by internet users. This can be done through text mining and semantic analytics.

There are many simple tools that allow you to be notified when there are discussions or news mentioned about a particular service, company or product. Some of these tools are IFTTT (If This Then That) where it can be set up such that the company gets notified when news about their company or service is mentioned. Another example would be to use data analytics platform like Google Analytics, Mixpanel and Kissmetrics that can let a company examine how many people has visited their tracking page or has tried to track their parcel. With such tools, logistics companies can analyse how often their recipients check when their goods will arrive, how long do recipients wait for the delivery to come before they check the tracking page, and at which stage of the delivery is the tracking page accessed. With such data, logistics companies can also use such metrics to improve delivery performance. An efficient logistics company would have fewer recipients obsessively checking the tracking page rather than a less efficient as long delivery periods or deliveries that pass its appointed delivery date lends itself to a more annoyed consumer who would try to track down their missing item(s).

All in all, Big Data is an area that remains largely untapped, especially in the logistics field, but its potential to add tremendous value towards the operations of the industry’s players cannot be understated. It is really important to note at this point that capturing all data with no goal in mind does not help. Just as it is crucial to capture data, it is meaningless unless the right data is captured. As margins tighten for the entire industry, a company adopting Big Data Analytics as part of their strategic management plan would surely be seen as a sustainable competitive advantage.

Share this post

Latest Posts
Stay updated with the latest news and posts by subscribing to us!