Tag: Jon Jackson

Guest Post: Returnless Refunds–Cutting Reverse Logistics Costs and Building Loyalty

Prof. Jon Jackson

Prof. Jon Jackson at Providence College raises an interesting logistics issue.

In the evolving landscape of e-commerce returns, major retailers such as Amazon, Target, and Walmart are increasingly adopting “returnless refunds”—granting customers a full refund while letting them keep the item. Though quietly deployed, this strategy addresses operational inefficiencies and builds customer loyalty.

For retailers, traditional online returns impose heavy costs: shipping back, inspecting, restocking or disposing of items, and managing the reverse logistics infrastructure. By eliminating the return flow, retailers cut reverse logistics expenses, simplify operations, and reduce strain on reverse-channel storage and processing staff. Many retailers now use decision-making algorithms to determine return eligibility, factoring in item value, customer return history, resale potential, and handling cost.

According to a recent study cited by the Wall Street Journal (July 24, 2025), the benefits of returnless refunds go beyond just reducing logistics cost. It can also encourage positive reviews, repeat purchases, and stronger brand loyalty—especially when the retailer frames the decision around convenience or sustainability motives.

Despite its promise, returnless refund policies must be carefully calibrated against the risk of return abuse. In 2023, it was estimated that customers returned $743 billion worth of merchandise (or 14.5% of the products they purchased). Of those returns, roughly 14% were fraudulent, costing retailers $101 billion in losses. If customers believes they will receive a returnless refund, it could lead to significantly more fraudulent returns.

In summary, returnless refunds offer retailers a strategic, cross-functional tool that enhances both reverse logistics (a topic in Chapter 11 of your Heizer/Render/Munson text) and customer experience. However, to realize their full value, they must be guided by data, aligned with brand strategy, and protected against abuse.

 Classroom Discussion Questions:

  1. How do returnless refund policies affect different parts of the supply chain, and what trade-offs must companies consider when choosing to implement them?
  2. Should companies be transparent with customers about when and why they are offering returnless refunds? What are the ethical and strategic implications?

Teaching Tip: The 15th Edition Ties AI Into Your OM Class

Prof. Jon Jackson

Our new 15th edition, just released, brings the topic of artificial intelligence into the course with AI in Action boxes and new material throughout the text. But we have gone a step further through our Instructor’s Resource Manual, a fantastic teaching tool. If you are new at teaching the course, you will find this 400 page guide an invaluable resource. Each chapter now contains an AI in the Classroom section, created by Prof. Jon Jackson at Providence College, providing 15-20 minute exercises. Here is a sampling from 3 early chapters:

Chapter 1
All firms, regardless of industry, can use productivity measures to track process performance. This exercise is designed to explore relevant productivity measures for different industries with the help of an AI-powered chatbot. Student groups can explore the following types of
facilities/firms (each group will pick one):
 Manufacturing facility  Market research firm
 Warehouse facility  Accounting firm
 Retail store  Financial services firm
 Restaurant  HR department
Students can use the following AI prompt structure: ROLE: I am a manager in a [insert facility/firm here]. GOAL: I want to measure productivity (an output divided by an input). REQUEST: generate 5 measures of productivity.
In groups, students can compare AI responses, evaluate the validity of the productivity measures (connecting to Chapter 1 definitions), and identify the best productivity measures to implement.

Chapter 3
A work breakdown structure (WBS) can provide a hierarchical description of a project into more and more detailed components. This activity is designed to practice this process for fictional projects around campus with the help of an AI-powered chatbot. Student groups can explore the following projects (each group will pick one):
 College graduation party  Student art exhibition
 Charity 5K event  Entrepreneur shark tank
 Intramural sports tournament
Students can use the following AI prompt structure: ROLE: I am the project manager for an upcoming [insert event here]. GOAL: I want to create a work breakdown structure to break the project into more manageable components. REQUEST: generate a work breakdown structure with 4 main tasks, each with 2 subtasks. For each subtask, also provide a short description and an estimated duration to complete the subtask.
In groups, students can compare AI responses, identify if any main tasks are missing (or unnecessarily included), and evaluate the accuracy of duration estimates.

Chapter 4
AI-powered chatbots can be helpful to enhance our understanding of confusing topics, but it isn’t guaranteed to provide accurate information. This in-class activity (15-20 minutes) is designed to get us in the habit of being critical of AI output, and if necessary, re-prompting the AI-powered chatbot to give a better answer. Student groups will try to answer the following questions with the help of the AI-powered chatbot:
 When does a 2-period weighted moving average equal the Naïve approach?
 When does the exponential smoothing method equal the Naïve approach?
 When is it best to use MAD vs. MAPE?
In groups, students can critically assess the accuracy of the AI responses (referencing material in Chapter 4) and identify more effective ways to prompt AI-powered chatbots.

For a desk copy of the 15th edition, please click on this link.

Guest Post: Rethinking Queuing Theory in the Age of Virtual Lines

Jon Jackson,  Associate Professor of Operations Management at Providence College, raises an interesting issue regarding waiting lines. 

Queuing theory emerged in the early 20th century with the rise of telephone systems and has since become a core part of operations management education. Classic models like M/M/1 and Little’s Law—staples of Module D: Waiting-Line Models—help us analyze everything from grocery checkouts to airport security.

But in recent years, a shift has occurred, one that challenges our assumptions about what a “line” even is. Increasingly, companies are replacing physical queues with virtual ones. Customers now “get in line” via app or text, receive real-time updates, and arrive just in time for
service. Disney’s Virtual Queue and Yelp’s Waitlist are a well-known examples, but virtual queues are also popping up in healthcare (e.g., Canadian ERs) and government services (e.g., North Carolina DMVs).

From a customer-experience perspective, virtual queuing offers obvious benefits: more flexibility, reduced perceived waiting time, and greater comfort. But from an operational lens, it raises a deeper question: are we still managing a queue, or managing something entirely new?

At first glance, queues—physical or virtual—follow the same logic: customers arrive, wait, and are served. But virtual systems change how that waiting is experienced. In physical lines, customers can see how many people are ahead, assess progress, and make real-time decisions about balking or reneging. In virtual lines, those cues disappear.

Virtual queues also alter arrival rates. Traditional models assume random arrivals and FIFO service. But virtual systems can shape arrival patterns via notifications and estimated wait times. This introduces a hybrid between queuing and appointment systems.

Fairness and prioritization are evolving too. In physical lines, order is usually determined by arrival time. In virtual systems, paid priority (e.g., Disney Lightning Lane) complicates this logic.

Should we optimize for efficiency or fairness—or both? Even foundational concepts like Little’s Law may need rethinking. If a customer isn’t physically present, are they still “in” the system? Ultimately, virtual queuing is more than a customer-experience improvement. It’s a meaningful shift that invites us to revisit historical queuing models and the assumptions behind them.

Classroom Discussion Questions:
1. In a virtual queue, does a customer “enter the system” when he joins the queue virtually or when he physically arrives for service? How does your answer influence how we analyze the system?
2. How does paid priority—whether in virtual or physical queues—impact perceived fairness?