The Science Of Shipping

The world of retail has changed quickly. Everybody wants instant gratification, and retailers are scrambling to meet those expectations — which is creating a massive burden in the world of shipping. Rob Taylor, Co-Founder and CEO at Pivot Freight, talks to MPD CEO Karen Webster about how those cumbersome problems can be solved with some scientific solutions.

As the volume of items that are being purchased online increases, so does the amount and complexity of difficulties encountered by suppliers in the retail space. These businesses face a slew of problems related to delivery times, tracking and customer service, and many are in dire need of a solution.

One such company offering a solution is Pivot Freight, whose Co-Founder and CEO Rob Taylor shared with MPD CEO Karen Webster his perspective on the potential for applying predictive analytics to shipping and freight handling.

Over the last five to 10 years, the rise of eCommerce in particular has really just changed the game for retail,” Taylor tells Webster. “And it’s not just for outbound customer delivery, but also for inbound supply.”

While what he describes as “a massive SKU proliferation” and omnichannel retailers deploying an “endless aisle” concept of product selection is great for consumers and retailers, it has placed a burden on the supplier network.

Retailers that had 500 suppliers five to 10 years ago might have 5,000 today. Managing that inbound and that drop-ship supply, from the standpoints of carrier selection, optimization and visibility, “has really become a pain point,” says Taylor.

Serious logistics issues like those can cost serious money to solve. As a result, retailers are having to re-architect their supply chains from the ground up.

Given the fact that a company can’t expedite delivery outbound if it can’t expedite it inbound, Taylor and his team at Pivot are focused not on the “last mile” of customer delivery, but rather on the supplier inbound and drop-ship. After all, as he explains, “that last mile is enabled long before, in the supply chain, and it’s broken all the way back.”

Taylor categorizes the two value propositions that a company like his can provide as “visibility” and “control.”

With respect to the control element, Pivot has a platform that connects into carriers on one side and retail suppliers on the other, allowing retailers to ship through it on behalf of their retail customers. When the shipment is entered, Pivot takes the unique characteristics related to weight, class, and the “from” and “to” locations, and pays the carriers in real-time based on the retailer’s negotiated rates with them. Automating and executing the shipment — rather than requiring the supplier to rely on static routing guides that Taylor calls “obsolete” — both reduces the cost for the retailer and takes the burden off the supplier.

As for the visibility aspect, Taylor explains that Pivot is able to dashboard the data running through its platform for the retail customer and show them everything in their ecosystem that’s in transit. The aforementioned predictive analytics come into play, here: If and when Pivot believes something is going to be late, it can flag shipments, giving the retailer the ability, says Taylor, “to proactively reach out to their customer and let them know what’s going on.”

Taylor adds that Pivot (which he says requires “almost no” integration on the part of the retailer because it is cloud-based) does not displace an existing transportation management system (TMS) that a retailer might already have in place. Rather, he explains, “we really enhance the functionality through aggregating all this disparate data…and putting some predictive analytics from suppliers on top of it.”

The shipping data that Pivot collects is currently being applied to building out two areas: real-time carrier selection and predictive delivery exception.

Taylor calls carrier selection for supplier inbound to drop-ship “a huge area of opportunity.” In running analytics to the shipment data on its platform, Pivot has observed that there is only 51 percent vendor compliance to retailer public routing guides. Furthermore, suppliers choose the low-cost carrier only about 28 percent of the time; had they chosen the low-cost carrier, they would have also seen a transit time reduction in 67 percent of those cases.

Simply put by Taylor: “It’s a mess.”

His conclusion is that, because the routing guides are not being complied with and the routing guide itself is not optimized, an ideal solution calls for automation and real-time decisioning — the application of which, Taylor tells Webster, has “consistently [produced] a 10 to 20 percent freight-spend reduction.”

The fact that the freight industry is “way behind” parcel carriers like FedEx and UPS in in its ability to inform consumers about shipments in transit in detail again, says Taylor, speaks to the need for predictive analytics.

One of the things that Pivot is doing in that regard is normalizing the data across carriers, and coming up with some standards for how things should be flagged.

The second thing — the “more interesting” one, in Taylor’s words — is that the company is “getting predictive around, in an automated way, when we believe a shipment is going to have some sort of exception on its path. Whether it’s a weather delay, or if it’s just sitting in a place that it shouldn’t be sitting for as long as it is, we’ll actually begin to proactively flag these shipments.”

As a result of those efforts, Taylor tells Webster, Pivot has seen almost double the amount of in-transit deliveries actually showing an exception — “12 to 15 percent” compared to the “6 to 7 percent” that are traditionally flagged.

Regarding Pivot’s plans for the future, Taylor says that the company’s objective right now is to really go deep with a limited set of customers” and learn more about the science-related problems that they are trying to solve. “We definitely want to be the de facto supplier inbound platform,” he remarks.

“The visibility piece” is really where Taylor believes the company has its greatest potential for growth. “Visibility can mean a lot of different things to different people,” he concludes. “For us, it ultimately means being predictive. So what we’re building here is a deep, data science, machine-learning, predictive analytics capability that we can begin to apply to lots of different supply chain problems,” whether they are domestic or international.