All That Goes Into Locating an Address for Last-Mile Delivery, and More

Indecipherable addresses have plagued the last mile for ages, resulting in lengthy undelivered lists and unwelcome challenges for delivery agents.

A hearty welcome to the 79th edition of The Logistics Rundown, a weekly digest that aims to put some perspective on what’s brewing within the logistics industry. This is a space where we religiously dissect market trends, chat with industry thought leaders, highlight supply chain innovation, celebrate startups, and share news nuggets.

E-commerce as a retail segment has shown remarkable consistency in growth over the past three decades, persevering through recessions and even a pandemic. The US has roughly 220 million people indulging in online commerce today, making up nearly two-thirds of the population. 

Understandably, this has seen the need for the last-mile jump over recent years. Demand for stellar delivery experiences pushes companies to dedicate a larger portion of their operational budget to servicing their last mile. This resulted in new delivery establishments opening for business and existing corporations' scale networks, attracting a sizable workforce into the last-mile market. 

At its core, last-mile delivery is pretty uncomplicated. The delivery worker's primary objective is to locate and deliver packages to the designated person and address. This isn't anything new; delivery workers have been doing this for decades without a problem. However, an age-old challenge continues to pose a significant obstacle in today's context, especially for the newly hired and inexperienced last-mile delivery workforce — the daunting task of handling wrong or indecipherable addresses.

Let's say you need to locate an address on the map. All you need to do is whip out your phone, type in the address on Google Maps, and you’d have the address located in a matter of seconds. Google would often help with prompts to the address even before you type it in completely. 

Now consider the address you're punching in is invalid. For instance, if the street name is wrong or isn't in the region you're searching, you'd notice the suggestions stop appearing. In a way, this indicates that something's not right with the address. While this does not do much, it does serve as a warning sign, telling you not to proceed with the particular address as there's an issue with locating it on the map. 

It does not work the same way if you're in the last mile business. For one, wrong or invalid addresses can get (and often are) typed into the system, and there are usually no checks in place. And once they are in the system, they can spread through into client information, shipping details, logistics, and all the underlying processes involved. Incorrect addresses will get replicated and come with unpleasant consequences. 

So how exactly is this a problem now, and wasn't talked about earlier? 

"When a USPS mail carrier walks down the street with a bag of envelopes, despite being pre-sorted, they still take a moment to inspect each address before making the delivery. They possess a keen understanding of the neighborhood, enabling them to recognize when John Smith's address is incorrect and he actually lives one door down," said Jeffrey Duncan, senior technical product manager at location data intelligence company Smarty.

"Consequently, they deliver the mail to the correct location without ever reporting the address discrepancy to anyone upstream. From the customer's and shipper's perspectives, everything appears seamless, all thanks to the resourcefulness of the last mile agent."

This information remains siloed within individuals, causing issues when a new delivery worker takes over the route or system changes occur. Expanding networks and a greater infusion of inexperienced workforce in the market has caused such challenges to balloon. While the problem has always existed and has demanded fixing, the conventional process has been to let the last-mile delivery worker figure out the address discrepancies on their own, as businesses often relied on their ingenuity and local knowledge to ensure successful delivery.

Duncan contended that some companies still depend on Google Maps today, leveraging it as a form of address validation. "But here's the thing — using Google Maps can give you an approximation of where an address would be if it were real, but it doesn't tell you whether the address is actually valid or not. It's more an estimation tool than a reliable validation solution."

Validating an address begins several steps before the packages are dispatched for delivery. "It's about tackling the ongoing problem of new addresses coming in, verifying addresses as they enter the system to effectively stop the bleeding. Once that's in place, the management must focus on the backlog of existing addresses. This is much simpler as customer interactions are no longer involved. The objective is to instantly notify the customer when they enter an incorrect address and prompt for a correction," said Duncan. 

Considering the different information silos baked into the system, it is also crucial to standardize addresses, removing improperly formatted addresses that could make it harder for delivery agents to comprehend. 

For instance, people use the wrong street suffixes — like calling it a 'Parkway' when it's actually a 'Boulevard.' And they could write improper abbreviations too, like 'strt' instead of 'st.' for street. Some streets have had previous names that have been changed multiple times over the years. Or multiple streets with a similar name that's only distinguishable based on additional address markers. 

"At Smarty, we normalize the address by rectifying suffixes and applying appropriate abbreviations wherever possible. For instance, 100 North Main Street might be written in formats like 100 North Main, 100 N Main, or even 100 Main. We ensure all these variations are resolved to the same standardized output," said Duncan. "Recently, I came across a single address with approximately 1,300 official variations of the same location. This included misspellings as well, but it's quite astonishing how complicated this could get."

The address trouble could go beyond just the street names. The US uses five- and nine-digit ZIP codes, with roughly 5% of all nine-digit codes changing monthly. While homeowners typically don't need to know their nine-digit ZIP codes, these changes are significant to the delivery system. 

"Obtaining client data is relatively straightforward, but acquiring accurate source data is challenging. We derive our source data from various sources, including the US Postal Service. Additionally, we have developed our own proprietary mechanism to gather data from external datasets beyond USPS. Our matching engine normalizes the input address provided by the customer and finds the best possible match within our dataset."

While the technology is relatively simple and the RoI is quite evident, the adoption curve with address validation solutions depends largely on the size and network scale of the company. Duncan explained that SMEs have shorter integration times compared to the likes of DHL. 

"SMEs are more proactive in seeking solutions and are willing to make phone calls to address their needs, unlike larger organizations that anticipate longer implementation times assuming the problem is more complex than it actually is," he said. "It's similar to restriping a small parking lot that can be handled by an employee versus restriping thousands of parking spaces, which requires procurement and a bidding process to hire external contractors. The bureaucracy and lengthy decision-making processes definitely delay integration."

For companies on edge, it would make sense to do a pilot with an address validation service, checking their record of missed or undelivered shipments to see if it can make an improvement. Routine validation analysis can help companies see how many addresses on their list can be corrected and how many would be undeliverable. 

"This information can then be used to proactively communicate with clients, informing them that their provided address cannot be delivered to, thus offering them a chance to provide an alternative or corrected address. By reducing the log of undelivered shipments, which essentially represents wasted money, companies can cut losses and improve customer experience by ensuring packages always reach the address they were intended for," said Duncan.

The Weekly Roundup

The National Retail Federation's Global Port Tracker predicts that US imports during this year's peak shipping season will be lower than in 2022 but will trend towards more typical pre-pandemic peak volumes. Despite continued consumer spending and projected growth in retail sales, retailers are anticipating a slowdown in demand after two years of explosive growth.

Persistently weak freight demand is prompting some LTL carriers to revise spending plans for the year, as they reassess capital expenditures after substantial investments during the pandemic-driven demand surge. While some companies are scaling back their capital expenditure plans, others are proceeding as scheduled, recognizing that investments made during economic downturns can yield significant gains once demand rebounds.

Inventory levels and tight supplies of diesel in the US are starting to impact spot prices in key markets, signaling a potential shift in the diesel market. While the average retail diesel price has been declining in recent weeks, physical ultra low sulfur diesel differentials in the U.S. Midwest have surged, indicating increasing demand and potential tightening of supplies.

A potential agreement between the International Longshore and Warehouse Union (ILWU) and the Pacific Maritime Association (PMA) is on the horizon after a year of negotiations, according to a Biden administration official involved in the talks. While the details and duration of the contract remain uncertain, businesses are cautious and are likely to continue keeping cargo away until the agreement is ratified.

…said who?

“It is a different climate than it was six months ago. And a number of our managers were surprised. Some of them had too much inventory on order, and then all of a sudden it got delivered, and people weren’t in the same frame of mind as earlier. Now we will start having sales when we didn’t need to have sales before.”

- Warren Buffet, CEO of Berkshire Hathaway, commenting on demand falling over this year and the challenges with predicting retail demand

---

Want to talk with us? Have something you'd like us to cover? Drop your thoughts to vishnu@truckx.com

We are TruckX, the Internet of Things plug to logistics. Check us out at www.TruckX.com