Venture Investing in Logistics Tech Startups: The Case of PAXAFE

Venture Investing in Logistics Tech Startups: The Case of PAXAFE

As Marc Andreesen said famously, “Software is eating the world.”

That statement is as true in logistics as any other industry. But in logistics, that trend has a unique twist. Many logistics tech startups are not located in San Francisco or Silicon Valley, but in the Midwest or Southeast where the customers reside.

As a venture capitalist who invests in Midwest companies, I want to use the case study of one startup we funded last year to illustrate how we look at venture investing in logistics tech. My fund, Comeback Capital, invested in PAXAFE, a startup focusing on contextualizing supply chain data to de-risk B2B shipments and enable dynamic cargo insurance.

At first glance, this does not seem like a promising investment. PAXAFE was entering a fragmented $18B visibility provider market filled with asset tracking solutions and visibility platforms. Providing supply chain visibility has become a commodity, especially if your perception of visibility is someone staring at a dot representing a real-time shipment moving across a screen.

But the lens with which PAXAFE looked at the function of visibility, and the approach and philosophy the company formed towards improving the quality of the data made supply chain visibility much less of a commodity product, and the investment opportunity very interesting.

A lack of real-time supply chain visibility leads to product theft, counterfeit, damage and loss through the supply chain. 

More importantly, not knowing the real-time location or condition of a shipment creates a lack of accountability and sub-optimization of inventory and warehousing labor. In particular, the lack of visibility and quality data from the supply chain makes it difficult to price shipment risk in cargo insurance.

Compounding the inefficiency and loss problem is a data accuracy problem. The $18B visibility provider market lacks a data standard. Solutions are inconsistent, unreliable and not contextual, which makes it difficult for cross-industry stakeholders, like insurers, to buy-in to the accuracy and validity of the data.

A lack of real-time supply chain visibility leads to product theft, counterfeit, damage and loss through the supply chain

Visibility providers are striving to provide an accurate prediction of adverse events –when, where and under which conditions adverse events are likely. If you do that, you can help shippers mitigate and avoid risk, while also helping insurers price and model for that risk.

Many companies seeking to solve this problem aggregate raw data from many data sources however, that data is not actionable because it is not contextual. If it is not contextual, it is difficult to form accurate and consistent prediction algorithms to understand how future excursions are likely to occur. If you cannot accurately predict, you cannot prescribe to customers how to best avoid specific risks.

We invested in PAXAFE because they formed an alternate hypothesis about the solution – focusing on data quality, not data quantity. Instead of relying solely on the aggregation of multiple third party and enterprise data sources capable of answering the commoditized ‘what’, ‘where,’ and ‘when’ questions, PAXAFE uses AI and machine learning to properly classify and contextualize supply chain data based on historical and real-time product-level information to answer the most critical inputs to the prediction of ‘how’ and ‘why’ did an adverse event occur.

PAXAFE recognized that visibility incorporates three core actions: intervention, resolution and prediction. They took an unconventional stance in determining that quantity and volume of data are meaningless if the quality of the data is not there; garbage in, garbage out. PAXAFE made the unconventional claim that accurate and consistent prediction will not be possible without proper classification and contextualization of product-level data.

Then they built the solution. PAXAFE created a contextualization platform that uses AI and machine learning to classify and contextualize actual excursion events. The contextualization platform feeds a more accurate prediction engine to identify when and where risk is likely to occur pre-emptively and enable companies to minimize said risk.

Weekly Brief

Read Also

The Evolution of Last Mile Logistics

The Evolution of Last Mile Logistics

Amit Prasad, Vice President, Supply Chain & Data Science, Coyote
Why Now is the Time to Choose LCL Shipping

Why Now is the Time to Choose LCL Shipping

Greg Scott, Director of LCL Ocean Services, C.H. Robinson
In a Time for Resiliency

In a Time for Resiliency

Erik Caldwell, President - Last Mile Logistics, XPO Logistics, Inc.
New Cargo Vessel Could Revolutionize Moving Freight on Mississippi River

New Cargo Vessel Could Revolutionize Moving Freight on Mississippi River

Mary C. Lamie, Executive Director, St. Louis Regional Freightway
Artificial Intelligence Explained.... with Cats and Dogs...

Artificial Intelligence Explained.... with Cats and Dogs...

Richard Bradley, Director, Digital Supply Networks, Deloitte
From Robotics to Big Data, What the Future Holds for Manufacturing

From Robotics to Big Data, What the Future Holds for Manufacturing

Karl Rosenblum, Head Global Capacity and Risk Strategy, Alcon