We kicked off our Germany conference season with E-Commerce Berlin Expo; one of Germany’s leading e-commerce events in February! With 7000+ visitors from all over Europe expected at the Expo, we were looking forward to meeting delegates, fellow speakers and retailers across Europe.
Our founder and CCO Dana von der Heide was also invited to share her insights on one of our key focuses for this year – machine learning and its applications for e-commerce logistics.
Problem: E-commerce logistics data is a mess for everyone:
The lack of standardised data and protocols affect the experience of all key stakeholders within the e-commerce logistics landscape.
For instance, unclear or incomplete logistics data creates frustration for consumers in the form of:
- Hard to understand delivery updates
- Unclear delivery times
- Incomplete delivery times
Merchants have limited visibility on carrier performance, affecting service quality and constraining the delivery experience they provide for their customers. Carriers struggle with limited visibility on the external factors and cannot adapt to consumers and retailers changing demands.
Fortunately, with the right data and services, there is a way forward to drive value for everyone involved in the e-commerce logistics ecosystem. In this article, we share some key takeaways from our Machine Learning (ML) presentation and how ML can be harnessed to create better outcomes for all involved.
1. Machine Learning is instrumental to predict logistics outcomes
Customer’s expectations are usually straightforward: orders should be shipped accurately and arrive on time. Studies show that 40% of online shoppers say they check an item’s order status at least once a day and expect up-to-date information on their delivery status.
Retailers who can accurately and proactively communicate delivery timing will provide their customers a significantly better experience than those who do not.
We at Parcel Perform use Machine Learning technology to help our customers configure predictions based on specified day range, prediction update sensitivity and how conservative updates should be. As a result, consumers receive a much stronger customer experience; with a more accurate delivery date predictions.
“Rather than just a location of the parcel, our (machine learning) model turns tracking data into a clear and reliable delivery prediction”.
How accurate delivery times be used to enhance customer experience:
- Sending accurate, proactive delivery notifications that is fully branded with the retailer’s colours and tone at accurate delivery intervals. Customers will feel reassured that the retailer is looking out for them and has their best interests at heart after checkout.
- When we expect delays in delivery, retailers can respond proactively with a clear, structured plan of how the issue would be resolved. For example, they could provide the logistics carrier’s contact number, invite them to get updates via social media or revisit order details.
2. Machine Learning can empower customer service teams to deliver high quality, proactive service
Did you know “where is my parcel” (WISMO) enquiries make up an average of 980 tickets per customer service agent? That’s a significant proportion of customer service time and resources that can be used for long-term improvements versus addressing re-active customer service issues.
Customer service teams which shift from providing reactive customer service to pro-active customer service can drastically improve their quality of work and other business metrics like customer retention and Net Promoter Score. However, to do that, they need to be empowered by the right processes systems and assets.
With machine learning, one can:
- Identify parcels with risk of an issue and suggest mitigation measures
- Automate notification flow depending on previous consumer interactions
- Determine when to communicate with consumers for maximum engagement
“72% of business decision makers say Machine Learning allows humans to concentrate on the meaningful work with customers.” (PwC)
3. Machine Learning enables improved efficiency and more accurate decision making for logistics managers
The exponential growth of e-commerce has led to immense pressure on logistics carriers to deliver better service at lower costs. Incumbent carriers are also now grappling with new, emerging competition in the supply chain industry.
These disruptive forces have provided a plethora of choice for retailers for their logistics operations. But it has also led to another problem: the difficulty to differentiate between carriers and the challenge to cater to varying customer needs.
Machine learning technology therefore helps to pick the right carrier for every parcel and enable efficiencies in the supply chain.
The right data and predictive models can help logistics managers answer:
- Can you make educated decisions about which carrier to use where and when?
- Do you know your actual real-time transit times? Are they meeting your expectations?
- Do you have a NPS for last mile delivery to optimize for?
Why not automate the monitoring and assessing with Parcel Perform so you can spend more time on other business tasks? Check out our Logistics SLA service available for all now. Find out more here.
We’ll next be at Inside Retail Live 2020 in Australia, where we’ll be exhibiting over the 2 days at Booth 28. Drop us a note if you’re also attending, or check out our conference guide here.
For a detailed consultation of how Machine Learning can support your business goals speak to our team today for a personalised consultation.
Joshua is the Marketing Manager at Parcel Perform. He loves technology and a good book. Outside of e-commerce and Parcel Perform, you can find him in the kitchen, gym or yoga studio.