Tackling Large Parcel Returns: How Data And Automations Cut Costs and Maximize Recovery.

Sure. No, think first of all, thank you for having me on the podcast. I've listened to quite a few episodes from you and they've all been very knowledgeable. ⁓ My personal motto ⁓ generally is to be able to learn every day and what can I do new? And one of the things that keeps me going back to work ⁓ is what will I learn new today?

And whenever there is an opportunity to connect with new folks like you or others, that is something interesting and obviously kind of ⁓ helps me get better out of myself. So I think that is one of the things that I kind of care a lot

Yeah, I resonate with that. mean, if there's no learning in what we are doing, it's not that fun. You know, when things start getting monotonous, it's super hard to keep yourself coming back to the same project, same problem again and again. And working at the scale that you are, I'm sure you're not short of any sort of problems or learnings on a daily basis. ⁓ So, I mean, I'd love to know more about you, Nikhil, with respect to your journey.

Where were you born? How did you start your career? And so on.

Sure. To start from the very beginning, I am from Delhi, ⁓ India. I was born in Delhi. I was brought up in Delhi. I did my ⁓ engineering from Delhi. I worked as a mechanical engineer, in fact, for three years for a company called Bechtel Corporation. I was involved in one of the largest refineries in India ⁓ and second largest in the world now with Reliance J3. Then I moved to US for my master's in industrial engineering and operations research.

And I happened to land my first job at Highmark in the product management world, which was by accident. And I think in 2016, we didn't have the title product management as such as popularized as it is today. So I think it was an accident, which turned out to be a blessing where I got to stumble upon the work that I am really enjoying and I really care about. It helps me keep myself on my toes from Highmark.

which was an insurance company. built single-member portal to help reduce touch points on the website. I then switched my ⁓ job roles from Highmark to Athena Health, which is also in the healthcare domain, but a slightly different function. ⁓ If you have heard of Epic, which make the electronic health record software whenever you go see a physician, Athena was competing with Epic in that space. ⁓

So at Athena, I was responsible for patient experience where one of the key problems that I'm really proud of that I solved was reducing no-shows cancellations and ⁓ reschedules in a healthcare setting. Approximately 40 % of the appointment end up being canceled, which result in two big problems. One, the patients who actually need to see a provider are not able to see them. Second, providers sit empty and they are

Mm-hmm.

one of the most heavy paid resources. So we want to utilize them as much as possible. That work entailed a lot of optimization, kind of opened my arena towards Wayfair, which where I'm working today in the optimization space. And we are solving some of the most complex problems in the supply chain world when it comes to moving heavy parcel, large parcel shipments from one point of the world to the customers in the United States, Canada, or UK. And we run into multiple challenges on daily basis, but

We are kind of continuously working on improving the customer experience, ⁓ ensuring that we are able to produce delivery dates that people can trust and reliably deliver on them. ⁓ That also led to kind of my final chapter, is also going, is my current chapter, which is consulting, because I realized working over there that this is not just a one-off problem, it's a large problem and a lot of people suffer because of it. So as a result, I started Silk Road Nexus.

which is probably with the goal of kind of helping organization or retailers to be able to manage their supply chain, manage their customers in a fashion where logistics is not a cost center, but more like a value driver. So that has been my motto and I've been fortunate enough to be kind of consulting and working with a lot of.

That's really nice. What was that shift from, you know, moving from healthcare to retail? those are two different industries altogether.

Absolutely. No, I think I love healthcare and I think I was solving some of the great problems and I think there are lot of emerging problems in healthcare and we continuously see them. I think one of the personal things for me is when I landed into this product management job was scale. healthcare, even though you work on problems, the problem set is still limited to patients or providers or healthcare settings within the United States. As a result,

⁓ the impact of your work kind of gets limited to a smaller audience. And as I was looking to transition and I was looking like I want to build something that is at a global scale, e-commerce was definitely one of the options because whatever you build gets shipped globally, impacts customers across multiple destinations and they have different sorts of problems ⁓ and different sorts of... ⁓

goals what you want to achieve. For example, a customer in UK may have a different need than a customer in US. So as a result, I was looking for a transition from healthcare and I think supply chain e-commerce ⁓ created that path or provided that path, in fact, to be able to kind of build that or solve problems at scale and globally.

Yeah, no, definitely. mean, the scale at which Wayfair and your operating kind of throws unique challenges with kind of meeting the promise that retailers and e-commerce giants have to make to consumers these days, especially in your particular space. I believe it becomes even more challenging given that you're dealing with heavy, heavy, the sofas and the TVs and the desks of the world. So in that particular paradigm, like where do you see

of, you know, like, how does a wayfare compare to any Amazon with respect to the supply chain problem and the delivery problem that they solve.

Sure, and think ⁓ it's not very different, to be honest. I think with the scale that Amazon operates at, it's enormous. think Amazon today is like a verb for anything that you want to buy. So it's become like a Google ⁓ for buying ⁓ options. ⁓ The key, I think, the differentiating factor between when it comes to a large parcel marketplace or a thing like Amazon is the difference in the average order value and the size of the item.

And I think those are very two distinguishing factors because if you look at Amazon's average order value, it probably is close to 80 to $100 or ⁓ even lower when it started selling more and more items. But when it comes to wayfair, our average order value is 300. So that kind of denotes two things that one, the things that we're selling are more expensive and people care about them more because the more you spend on something, the more closely you attach to it and a more expectation that you start to have from a retailer when it comes to

quality of the product, price of the product, ⁓ availability, and then finally how it's handled

those are like the two paradigm shifts that happen between, you know, how Amazon operates and what Wayfair does. ⁓ And I think that's what distinguishes these two companies in general.

Got

it. And in the same thing, do you think also having to deal with larger parcels, there is an optimization problem, I'm sure like that, but the one that you're dealing with where smaller parcels can easily fit on pallets and can easily be transported from point A to point B. How does it change when it comes to the sofas and the beds of the world ⁓ where you're dealing with ⁓ non-standardized dimensions for different products that are getting transferred?

Sure, and I think I would not call it non-standardized. I think the dimensions are standard because we work with suppliers to ensure that the dimensions are standardized to an extent that we know what these dimensions are. But when it comes to a difference between a small parcel or a large parcel item, ⁓ know, small parcel have a different problem altogether when it comes to shipping and logistics, which is more like ⁓ how do you ensure, first of all, the damage rate

remains low and then second, how do you continuously ensure speed with Amazon? The economies of scale are so big that they're able to achieve that. And I think that's one of the biggest driving factor. Plus, I think they have built technology from ground up for all of their logistics supply chain, which provides them great visibility into ⁓ their shipments. And then they are able to build technology on that visibility that allows them to kind of build the next level of customer moment features.

But when it comes to large parcel shipments, it's very different. First of all, lot of these shipments and ⁓ these items, you cannot forward position because of the size of these items. So you have a limited space. So you have to very carefully analyze what items to forward place in your inventory. Second, think you have to very carefully, again, design your network as to what items you want to drop ship versus ship from your forward position inventory.

you have to have that relationship with your suppliers where if you're selecting their dropship ⁓ origin as the starting point of the shipment, they are reliable enough to meet the lead times, they are reliable enough to put the items on the right truck or the right shipment goes into the right trailer. Those are different kinds of challenges as you kind of move from sizes of the shipment. and I think those are...

different problems with different dimensions. And Wayfair, I think, has been able to evolve across that because that's their primary bread and butter now. And we have built that relationship with the suppliers where we can understand ⁓ what products we want to ship from there and how we want to manage the shipment across different kinds of carriers. ⁓

Makes sense. mean, that makes sense, given that you have done the hard work of kind of working with the suppliers itself to kind of know what dimension, what kind of products are going to be fitting into different containers. It has now become more streamlined across the rest of the supply chain with respect to already knowing what to expect when the goods are moving ⁓ from the factories to the ports, to the distribution centers, the fulfillment centers and whatnot. ⁓

Does it change when it comes to the returns? Does anything change or do you guys have the same... Not just you guys, but how do you think it changes on the return side of things where goods are coming back from different consumers? Does the same problem or does the same paradigm work? Or do you guys have to change something about how you think about this?

No, returns is a completely different ballgame when it comes to large parcel returns. Because, for example, starting with small parcel, mean, what Amazon has ⁓ famously done is that it allows customers to return within 30 days. You have a label that you can print and then you can drop it off at a UPS or a FedEx near you. And then FedEx or UPS manages that return back into Amazon's fulfillment center or returns collection center. And from there, it gets sorted and it ⁓ gets rated as resellable.

refurbished or liquidation kind of inventory, ⁓ which is a model that can be adopted when you have easy packageable items, ⁓ understandable dimensions of a product, and you have built that trust with the customer that you can buy as much as possible and return that back if you don't really care about that product or you don't need that product. But when it comes to large parcel, there are different nuances. the customer standpoint as well,

the interaction with the product is completely different because you won't buy a sofa or a table or a carpet just for the sake of seeing how it looks. You probably want to buy that particular product to keep it for a longer duration of time. So the expectation is that the product should meet what it actually showed on the website or how it actually felt on the website. ⁓ And from there, ⁓ when that expectation has happened, when the product arrives, let's say in

⁓ not in a perfect condition, which could be damaged, ⁓ which could have a different feel than it would look like on the website, customers tend to return them back. But now comes the challenging part as to how you make that happen. Once you unbox a bed or a sofa, it's very difficult to put it back together. So now you cannot just expect a customer to have the same packaging as it came in. Versus if you remember, if you have returned products with Amazon, it's so easy that you have

let's say a small item which came in a decent size box, you can just put it back, wrap the box and send it out. Now with beds, it becomes really difficult. So now it's an expectation that is developing from the Wayfair side on the customer to be able to repackage it or put it in a situation where it's easily collectible ⁓ or it has to be de-assembled in such a manner that if someone is coming to pick that up, they're able to put it in a truck.

So that's where the challenge is very difficult. ⁓ And it requires a lot of ⁓ customer training as well to certain extents. And that's why if you look at it, a lot of these retailers who offer these large parcel shipments sometimes don't even accept returns because it's just so expensive to manage them that it's easier to eat up the cost of the product itself from the wholesale standpoint and the initial shipping. So it becomes a real challenge. And once it reaches the warehouse somehow,

Mm-hmm.

then you have to grade that inventory, which is another operational cost and burden. It takes up space because this is large inventory. It's already open. So you have to decide what to do with it. So the decisions have to be really quick to really capture value out of that returns process, ⁓ all in all, from customer initiating the process to a retailer or a marketplace, getting it back and then deciding what to do with it, like send it back to the seller, resell, refurbish, ⁓ or repair, or kind of liquidate.

That makes sense. I I can totally imagine sofas and beds being a different kind of return problem as you were describing it. So when that is happening, as returns are coming, I think it's very important ⁓ as a retailer, as a brand to make sure you provide them all the information upfront about the product so that the probability of a return is very low. Returns sometimes happen because the manual was missing, a nut was missing or

It said it was this dimension and it turned out to be a different dimension. also know like Wayfair and a few other home goods retailers are experimenting and I've actually scaled it the augmented reality or the augmented vision AI where you can actually ⁓ project a particular product into your space that you're trying to put it into and see how it would look like before buying so that you're able to reduce that return. So all in all, think the best way to save returns is not to have returns.

how to make sure that you can give a consumer as much information as you can so that the probability of them returning is low. even when that happens, I think as you were describing, it's super hard. Like taking it back itself is going to be, you know, like it's going to take the cost of shipping a hundred pound or a 50 pound desk or a shelf itself is so much that why even bring it back? So in that paradigm, do you guys, or do you look at it as a leader, product leader, look at it as

What is the role of data and AI in it? How to solve this problem starting from a customer returning a thing? Where do you think retailers and brands should leverage data and AI to be able to solve this sorting problem?

I think it's

very interesting and I think it could be tackled on multiple fronts. ⁓ As you said, return could be because of multitude of reasons. Let's say the easiest reason that we can encounter is that it is missing a couple of nuts and bolts. ⁓ Then you start to understand, I need to solve a problem where customer doesn't want to return a product because it's damaged or they don't want it, but they want to return a product because it's just missing a few parts. You start to look at that problem from a replacement parts angle.

So you ask customer questions like, hey, ⁓ why do you want to return? And if the answer comes back as, hey, this particular product is missing two parts, and I don't know where to get these parts. So instead of making a return, you enable replacement parts much easier. And to make the ⁓ replacement parts easier, you enable it on the UI. You go to the UI, you say that I want to return a product, and then

Hmm.

You just answer a few questions and then say, OK, our replacement part is required for these two products. So you just ship those two specific replacement parts to the customer. So everyone is happy in this particular paradigm because customer first, they didn't go through the hassle of repackaging or seeing what to do with that product. Retailer like Wayfair saves money and cost and energy in terms of not making that happen. And seller is also happy because now they only have to ship two retailer parts.

which we can do easily through FedEx because now we don't have to use a large parcel carrier to make that happen. But that's the easy part of the problem or that's the part where you can solve that problem. But multitude of things can happen where you don't expect to happen, which is customer actually didn't like the product. They actually want to return it. They didn't like it. Or the product arrived damaged. You can have replacement part for damaged.

specific items when it comes in a DIY situation. But some of these items, like if the headboard is damaged, shipping a headboard back to the customer is still very expensive. So in those situations, you start to analyze, use data and AI to interpret what kind of products you are actually seeing that happen most on, ⁓ what kind of suppliers are actually the biggest culprit where the returns are happening. And then you start to identify

Mm-hmm.

⁓ lanes or product types that actually get damaged during the logistics part. So for example, right, vanities is a very good example where they get damaged very easily during transit. And the only option is a return of the vanity when it arrives damaged to the customer. You can't replace, do a replacement part order. So when you know from data that vanities get damaged, what we do is we open the vanities in the last mile, check if they are undamaged or in what condition they are and repack them together before shipping again.

to the customer. So you're making sure that ⁓ you have identified explicit products that can get damaged during the supply chain. You have identified suppliers, are culprits. And you have also identified what kind of product categories are highly likely to be damaged in certain conditions. So you make those kind of transitions on the operational stages using data to be able to vet them and ensure that you deliver them in a good condition or an acceptable condition to the customer.

Makes sense. It's like analyzing the data, finding out upstream where exactly the culprits are coming from and kind of solving them. And how do you kind of implement this on a warehouse floor ⁓ particularly?

Yeah, I think that's where the challenging part is, right? Because now you have to introduce new operational processes which have not been part of ⁓ the regular operational processes which are in tuned into your muscle memory. So you definitely need to evolve those operational processes over time. I think operations is one thing which is never constant because you continuously improve, add new processes, you're launching new programs. So you have to continuously.

⁓ bring the ⁓ operational team together to be able to deliver on those milestones. And I think the ⁓ challenging part is not just bringing them together, but ensuring that these processes are followed. So you have to build metrics in such a way that track how these processes are getting followed. So first definitely is the operational process that you want to build. Second is metrics around

how these operational processes are getting implemented on the ground and what is the actual outcome ⁓ from these metrics, and then derive your circular vision again and see what needs to be changed and how it needs to be changed. So it becomes like a circular process of process, metrics change, process, metrics change. And that's how I think we have been able to achieve a lot of.

Correct. So it's like keeping an eye on whatever the SOP has been directed, has it been followed through metrics? Because the scale which you're operating at is like multi-facilities, global operations. You have to tie yourself to metrics to able to see that whether the SOP that you put together using data and ML and AI has been followed to what extent and what those bottlenecks are. how much does the human adoption aspect come into it where...

Sure, there is an SOP, but however you're dealing with actual warehouse workers who are used to working a certain way and having a certain kind of mindset when dealing with problem, right? Everyone knows better in some way or the other. So how do you kind of explain and solve that human like worker adoption challenge with respect to bringing in new processes around returns or in general, even in the forward flow?

exactly.

Sure, and I think I would like to break that into a couple of factors because it's not just the adoption problem of humans. think it's first of all, it's a big labor problem as well because now you're adding new processes and you already have an existing volume to take care of for the forward moving logistics. So you definitely are short on labor. So you first need to understand what your labor demand is going to look like ⁓ and how you can automate a lot of these processes. So first is labor understanding, labor demand understanding and how much labor is required.

Second is how much of these processes can be automated. So lot of the things that ⁓ require human inputs, for example, scanning and visibly checking if the product has a damage or not, you can automate those processes. Like you put a QB scan machine around the product and then just X-rays it and then says, OK, these are two areas where I see a damage. And then you have someone from a human level come in and interject and say, OK, is this damage repairable or not repairable?

So you definitely have to work through a mix of human and automation to ⁓ smooth in that particular process. And at the end, to your main question, what is the challenge with the adoption? I think it's not a challenge with the adoption, per se, that people are definitely willing to learn and people are definitely willing to put in that extra effort to work through that process. I think it's just the expectation setting that is the biggest problem. Because if a

you are working on the operations floor or a warehouse floor, the biggest challenge is you're doing multitasking. You are working on inbound, you're working on outbound, you're working on pick-pack, and then at the same time, now you have added additional processes on top of it, which is to ⁓ check return. So I think the expectation setting around ⁓ what is your role on the floor is the key when it comes to human adoption, because I think humans are very good at adopting. It's not a challenge that you face from a perspective that...

I don't want to do this. It's more like I already do this, and you want me to do this as well. So making sure that you have a clear path towards tasks as to what a person needs to accomplish has to be very clear and, in a way, ⁓ incremental to their journey as well. They don't want to go and do a work that a lower level associate does or even a lower level associate to them does. So they want to be incrementally going upwards when it comes to

human nature. So you have to continuously make sure that they evolve from their tasks and clear expectations around it. So change management becomes easier when you have that kind of expectation setting available to

Yeah, makes sense. having human incentives, like what is in it for them to be able to follow the process? It could be, it would save you X amount of hours later down the line in terms of maintaining your warehouse, keeping your warehouse up and running, or giving them incentive in terms of if we excel at this, we would save the company X amount of dollars, which would obviously reflect well on their resume and their ⁓ career at the company. think, yeah, I mean, I totally agree with it.

cannot be just like, hey, we are doing it without explaining why we are doing it. think when you need human adoption, need to explain in much detail what is in it for them, why it matters to their job, and how it would help them grow and their warehouse kind of operate at the scale that they want to. In this particular paradigm of just returns, where do you think cost comes in play? Because at the scale that you're operating, how do you ensure? I'll start with the start act.

And then I think we can double click on it. when a return comes into a warehouse, takes about 66 % of the cost of the retail value of that product to handle that return. that's like two-thirds of the cost. So if it's a $300 item, can cost up to $200. Actually, the case of I think the beds and the sofa, it might not be that much. But at an average level, 66 % of the retail cost of a product goes into processing a return.

So how do you guys look at that data? What needs to be touched? What doesn't even need to be touched to be even opened? Because that itself is cost, right? ⁓ So yeah, that's my question. Like how does cost play a role when returns are getting processed?

Now, think ⁓ first, I am not 100 % sure about what percentage of our cost is actually ⁓ driven by returns in general in terms of percentages. But I can give you a general perspective around what really drives cost and how can you minimize that in general. If that kind of helps answer your question as well, think, at the end. So when you look at the returns value chain in general and the journey, when it starts from customer initiating it,

Yes, okay.

to you receiving the item, to you inspecting the item and deciding what to do with it. I think all of these aspects have cost associated with it across this value chain. ⁓ And that cost may be tangible or intangible at certain points in time. Like you can feel that there is cost associated right now, but then there could be an intangible portion of that cost, which is linked to, let's say, customer loyalty. The customer might not come to your website again to buy the product if they are not happy with the process altogether.

So that kind of also drives the cost factor because if CLV goes down, your cost is much higher because you're losing future sales as well. So when you look at the tangible portions of it, mean, the tangible portions as you rightly said, right? Anytime you touch a box, a cost is associated with it. Every touch is costly. So one is how can you minimize and how can you fast track the process?

To minimize the touch process, automation comes in handy. Because if you can automate a lot of these processes, which is inspection, grading of the inspection, and then deciding what to do with that particular item, that kind of saves a lot of money because you are now reducing touches ⁓ on the returns process. When it comes to time, time is actually very crucial because a lot of the sales actually drive through peak seasons, which is the holiday season coming up right now. A lot of people buy ⁓ items.

Now, if you buy certain things that, let's say, are seasonal and have to be sold out in that particular market frame, which is the Q4 of ⁓ US holiday season, then you have to be very quick in terms of analyzing what to do with that product. And any extra hours spent on making that decision what to do is basically a cost associated with it. So in general, to be able to avoid those costs, you automate a lot of these processes. Second, make faster decisions.

And then third, when the decision is made, move that inventory to that location as quickly as possible. If it's liquidation, move it out of the warehouse because you're in the peak season already. You want to make space for new orders coming in. And if it has to go back to the seller, make that decision fast enough so that if seller has to make amends, they can do that as well. And if you have to sell that as an open box, you put it on the website as well. those are the things that you have to kind of make sure to continuously happen, which is automation.

⁓ customer loyalty doesn't get disrupted so that your CLV doesn't get impacted and obviously time to process all of that is very key.

Yeah, no, I mean, that's very insightful. Like you do have to accept returns because consumers would stop coming to shop with you because now returns and having that flexibility, if something doesn't work out as expected, it would be returned. So my money would be returned. So I think from that perspective, it's the cost of business. Now, from your perspective, you just need to optimize and automate as much as you can in the warehouses so that that cost of ⁓ maintaining that expectation is as low as possible.

So is it very similar to how, you know, so at Wayfair you have optimized the fulfillment orchestration by 48 hours. Is there a similar paradigm where, you know, like even returns can be optimized by some amount of time that hey, the decision making of where it needs to go. Does it need to go to the donations? Does it need to go to the liquidation? Does it need to go to back to the shelf or does it need to go back to the seller or slash vendor? Like, is that sort of like similar to how fulfillment is orchestrated and optimized?

Is there a need ⁓ or maybe the similar project still exists or already exists with respect to how to optimize returns?

Yeah, think optimization and orchestration, would kind of segregate these two. I think returns again is a very metric driven business and it has to be very metric driven as well because when you talk about costs or time, you have to measure each and everything and see how you're doing on those metrics. And when you start to drive your business based on those two metrics, you start to make decisions that could help you both on cost as well as time perspective. So then the question becomes once you are able to identify what

lever you want to drive the most, ⁓ depending on the nature of the business, the key is then optimizing for it. So if it has to go back, ⁓ you need to be able to identify ⁓ optimized path to move that back to the location where it's supposed to be. Then again, requires the complete optimization, which I don't think is a problem for ⁓ a company like Wayfair because we already have processes existing. The challenge again comes in the operational process because a lot of these

forward bound trucks are not going back to the same warehouse. It's always forward, forward, forward in warehouses. So you specifically need ⁓ backward logistic trucks that take it from point B to point A. ⁓ And that becomes a challenge because returns, you cannot predict how much is going to happen in a particular quarter. You can definitely come to a ballpark, but again, the volume is very variable.

So you end up having a lot of returns items stored in a warehouse for an extra day or two because you don't have enough trucks taking it back. So, and you don't want extra trucks because if you have extra trucks and you're basically paying for an LTL carrier or a full truck load carrier for items that don't exist at that point in time. So it becomes a very balanced ⁓ approach as to how much you can hold and you have the capacity to hold versus how much you want to ship back.

So the optimization from that perspective is challenging because you have ⁓ limited insights into ⁓ first the dimension of the product because when you optimize, you need to understand what is the dimension of the product because now it's open box. You have limited understanding of how the dimensions are going to play a role and then how much volume you can ship back or should be shipped back. ⁓ From a decision perspective, I think it's very quick because now you have the metrics that kind of tell you where the

⁓ what levers you want to pull from a business perspective so you can make faster decisions. But from a technology and operations standpoint, there's still a gap that exists in terms of ensuring that you take, once the decision is made, you are able to execute on those decisions. So that gap still exists. And I think over time, you have to kind of work through to close that gap by bridging operational processes much closer and kind of mirroring those processes very

very diligently into the optimization or orchestration software.

That makes sense. that same thing, but do you think, operating at a scale of Wayfair or an Amazon or a Walmart, where the kind of products that are getting sold are millions and millions of products? How do you standardize across and classify across this happens to this particular product based on this product? Is it more human-based decision-making or is it more automation-based? And even automation, I don't think, has reached a point where it can classify among

You know, you know like how a lamp is different from a different lamp a brand a different brand can have this as a grade a Return to vendor it could be you know, liquidate or you know donate and so on So how do you kind of look at it? Like when you have a maybe let's just say five million five million SKUs in your in your cat in your in your selection and Things are coming back from all over the place with different categories, you know, you you never know what's coming back to you in that sense

It can work if you're just dealing with single SKUs and you know what to do. at the scale of a big box retailers, how does it kind of change in terms of grading and routing?

No, it's definitely a challenge. I think that's where a lot of things depend on, think, and I don't manage the returns, so I will kind of speak from my position of how I believe it should happen and what really happens. In terms of ⁓ managing those, I think there are two critical levers. One is your supplier contracts, because when the items come in, you know what suppliers provided that particular item. So your contract around

are they willing to accept returns or not, and what condition returns are accepted. Because if something happened during the transit and supplier says it's a big box retailer's responsibility to manage from that point on and we will not accept a return, then it becomes a decision around that retailer to decide what to do with it. ⁓ And once you know that SKU and you know the options that are associated with that SKU, so you can definitely. ⁓

try to sell it on your website as an open box. So I don't think that's a problem in terms of managing the expectation of what product or what product class it is. I think the data is present over there. It's just about ⁓ ensuring that you have enough grading and good supplier contracts to understand what to do with that product. When it comes to grading, it's obviously a human decision. Certain humans can grade a same set of damage as grade one or a grade two class.

So it definitely becomes a very human intensive process. ⁓ And I think there are ways to standardize that as well. You have SOPs that a human can go through, but again, it leads to interpretation to certain extent. And that's the best you can do at this point in time if you don't have enough automation or automation at a level where it can take a human level decision to decide what level grade it is. ⁓ And once you have decided what level grade it is, you already have your wholesale pricing.

You know how much loss you've already made on shipments. So you can make a very ⁓ deliberate call or a cognizant call around what should be my pricing when you sell it as an open box. So managing product class and inventory, don't think is a challenge for any big box retailer because of the availability of back end data, you can quickly match it. But I think the problem comes down to supplier contracts and grading level as to decide what to do with that inventory.

Yeah, no, 100%. I think the grading level is the challenge which even retailers have a hard time figuring out because when you're dealing with five million kind of, know, like millions and millions of SKUs, no matter how much grading, every time you'd come up with a new problem and a new kind of human to have to make a decision. And that would kind of play into what exactly happens to that product at the very end. Where do you think sustainability and carbon plays a role in it? Like, how should retailers and brands think about?

sustainability in general when it comes to the returns problem. Should they be taking that just much like we are taking cost, space, value into consideration? What role does sustainability and carbon emissions play in that?

I think I'm very pro as well in terms of carbon emissions and sustainability. We have to kind of make sure that this planet survives for our grandchildren as well. But I think retailers take that seriously as well. now specifically with government regulations as well around EPA and emissions, ⁓ it has become a very cognizant effort to ensure that once you start moving items, you are cognizant of the fact how much pollution that you're creating, not just through

running trucks, also through liquidating certain kind of inventories or what kind of inventories you are liquidating. So that becomes a very cognizant goal as well for the retailers. And I think it definitely plays a role in terms of making the final decision. ⁓ know, one of the factors is that, know, with government pushing down the lot of regulations to be more green friendly, there are also secondary buyers that are green friendly and want to be green friendly.

So it becomes kind of a good market as well when it comes to returns to sustainably sell those products in an open market for people who are also looking for sustainable products and are willing to pay, let's say 10 % higher than the actual liquidators or other vendors to buy those liquidation inventories. So it definitely plays a factor. And I think it is one of the core decision makers as well when it comes to what to do with that inventory, how much pollution are we going to generate when you move that inventory.

versus completely liquidating it.

Yeah, no, % retailers need to think about how can they give that product a second life and the third life and a fourth life versus just the first life. And I think most big retailers, including Wayfair, are thinking along those same lines of how to introduce the emissions and ensuring more products are in circulation versus because the returns problem has been increasing for a while since COVID and even before that. And now it's at a point where if the

if the returns itself are not like if the value out of them are not maximized, that's a financial impact as well as with all the regulations coming in, it's super hard to kind of justify things going to landfill. Awesome.

Yeah. And

I think it is also about building that circular economy, as you mentioned. I mean, how much of that product can you refurbish and resell so that you minimize the carbon footprint? Again, how much of that ⁓ product can be used to build other byproducts, ⁓ which is a good part of circular economy? Why do you need to completely put that product into waste if you can generate byproducts from it ⁓ or create subproducts from it? Why not just do that? So I think that those are.

good factors, not just from an environmental perspective, but also from cost and value perspective for an organization because ⁓ you kind of have to come up with ingenious or out of the box ways to be able to understand and curb costs when it comes to return. circular economy with refurbishing and ⁓ circulating as byproducts definitely, I think, has a good future and people will start thinking in that direction soon.

Yeah, I think that's actually a very good point. I was reading a book as well on it. Like, how do you kind of use the raw material that is coming back from the returns ⁓ back into kind of circulation with respect to maybe it's a dining table plus five chairs, maybe one is broken, that doesn't mean the rest of them are bad. And as a single unit of SKU they don't need to go in a landfill or anywhere per se, they need to stay in the warehouse so that next time the order comes or a replacement comes in, you know, these can be reused.

Exactly.

as well. It's hard too. It's hard again. I think it needs a cultural shift in the industry where people start thinking like that because right now I think ⁓ the focus is not that. However, I think there some, ⁓ there's a good use case for it, especially with, you know, like reshoring of manufacturing, I think it would even make more sense so that more material is available within America to be able to do things. So yeah, I'm excited to see what holds, I think, ⁓

So clarity itself as a topic as well as a practice has been ⁓ on a game, but there's a lot more that we can do as an industry.

Exactly. No, I think it has to be a very conscious decision and a push decision from the leadership side as well to kind of make sure that everyone works towards it. But I think it definitely makes sense from where we are in the industry to kind of start looking into these avenues as well before it's too late.

Yeah, I mean, we as an industry have optimized forward flow to the teeth. Like we have actually built so much cutting edge technology to make sure things move forward as soon as possible, as cost effective as possible. I think if we use that same talent and same mindset to kind of solve the backward flow and the circularity flow, I think this problem is completely solvable. It's just, and also profitable when actually executed at a pilot level and seeing that, all right, what's the ROI to the business?

The other ROI, think that's like you were talking about customer loyalty, right? Like our consumers are changing as well. Gen Zs are going to be buying, Gen Zs are the ones who buy new, are going to be buying new homes, setting up new apartments, buying new products just to set up their life. And those folks and that particular segment of consumers are very sustainability focused. So it also reflects good on a brand, on a retailer or anyone that, I am pro sustainability. I am pro circularity.

I am offering you goods that otherwise would have impacted our planet. So it's also a good customer loyalty strategy to be circular in your operations because that way you're acquiring these customers who are otherwise going to go Goodwill or other thrift stores versus the big retailers in the world. it's not just a...

Yeah,

and I think to that particular point as well, you said about customers are already conscious and they want products that are more sustainable. I was reading a report, I think from McKinsey or Bain, someone that a lot of the buying emphasis is actually where they can see in the description if this product is sustainable or not. And people consciously buy those products, even though they have to spend 10%, 15 % extra for a similar product.

but they would spend that 10 to 15 % extra just to kind of get into the dimension from the good of the heart to be able to buy more ⁓ carbon neutral products.

Yeah, no, 100%. Well, Nikhil, this has been a pleasure. I have a lightning round for you. ⁓ All right. So just one word, like kind of giving your thought on it. It's a mix of personal questions and professional questions. So I'll start with the professional ones, and then I'll close with the personal ones. ⁓ So one product principle that you follow ⁓ while building supply, while solving some messy supply chain problems.

how it's going to impact the buyer.

the customer you mean.

The customer. I mean,

customers could be interpreted in multiple ways. That's why I use the word buyer because customer could be an operational stakeholder as well. But buyer ⁓ slash customer who comes and buys a product. If I'm solving anything, I need to ensure that they get what they need when that problem is solved.

100%. One myth about automation in supply chain.

overrated.

Yeah, yeah, I think you you cannot actually remove the entire human element out of supply chain as much as you know, like marketing and know, robotic companies, companies would want us to believe. ⁓

Yeah. I think

what it's going through is that particular cycle where we are overestimating its impact in the short term and then underestimating its impact in the long term. But I think in the long term, where the automation might go is like your warehouse may fully be automated, but you will still need ⁓ human hands to support that automation. It might not completely take away those humans.

Mm-hmm.

Yeah, definitely. think there would still be an element of human intelligence required when you're dealing at the scale. If you're dealing with one SKU you're sure. You can do it. at the scale which almost all retailers deal with, it's super hard.

But if you're dealing with one SKU, you don't need automation. One human hand is sufficient. That also is a connect run, like when to apply automation.

Yeah.

Yeah,

definitely. One metric that you almost look at every day for your business.

⁓ conversion and reliability index both.

Got it, got it. That's awesome. All right, moving to the personal ones, favorite book.

Favorite book, that's a tough one actually. That's a tough one.

I think one I would say is I would give two. One is the book called Noise by Naseem Taleb And the other one I would say is Originals by ⁓ Adam Grant. Both rank very high. I think Noise has helped me cut down and focus on things that matter and not worry about 100 things that go in the mind. Originals is something that I really like where

the comparison is that when you're an original, you're not building something original, but you're trying to do it in a different way than what has traditionally been done. it's about how to go out of traditional ways and look for opportunities that could be presented to you once you start looking at things from a different angle.

Yeah, the second one I ended up to read, especially because I resonate with that particular principle. You don't need to reinvent the wheel. You just need to see how legacy systems, processes have been working and seeing where exactly the improvement can be because sometimes reinventing the wheel is also going against the nature of how things have worked. And not everything like that works. ⁓ Last question from me. ⁓

Many individuals start their career in product and supply chain or in general, like either of those two fields. What advice you would give somebody who's right out of grad school, maybe they did a BS in supply chain or are aspiring to be a product manager. What piece of advice you would give somebody who's starting out right now to build a career in supply chain.

I think one thing that I have seen and I've learned from it is there are two parts that you can take ⁓ in any industry. One is a generic part, which is I'm a product manager. I can go work in healthcare, finance, e-commerce, and I can solve any kind of problem. That is definitely one of the parts that can be taken where you become a generic, you're a problem solver. Then the second part is like you become specific to one industry. You learn the ins and outs of that industry.

and you try to grow in that. I think I have now come to the conclusion, at least for myself, I don't know if this applies generally. So I don't think this should be taken as an advice. But at least what has started working for me is to build a niche in a specific industry. Because what it does in today's world of AI and hype around AI and problems that AI can solve, I think one thing that AI can't do is replace if you have deep expertise and knowledge in a particular domain.

you would always be ahead of AI, even though AI can reason equal to a human. But it can't build that level of expertise and human connection that you have been able to build in that particular industry. So initially, was on the path of like, could be product managers could be generic. But over the years, I've come to the path that being specific actually helps you grow and grow faster and be more knowledgeable in a space. So my advice would be if

You could be on either path, but you have to make a ⁓ very diligent call and an intelligent call from your inner side what path you want to take. But if you choose one, stick with that. ⁓

That

makes sense. That's really awesome. Nikhil, I really appreciate your time. This was a pleasure having this conversation. Where can people find you and connect with you, specifically to get advice or with respect to the Silk Road? ⁓

Sure, think the easiest way would be LinkedIn. Connect with me on LinkedIn. ⁓ I think we can put that in the description down below with the video. ⁓ And then other than that, on my LinkedIn, you can also visit my newsletter, is the Silk Road Nexus, also a consulting company. So if you have any questions or doubts, you can definitely connect with me ⁓ either way, through LinkedIn or through Silk Road Nexus website.

Yeah, we'll put all of that in the comments below. And yeah, I highly recommend anyone in the supply chain industry reading it because it's coming directly from somebody who has solved the problem versus just chiming on their mom from the outside. So this has been a pleasure, Nikhil. I really appreciate your time and thanks a lot for listening, everyone. Until next time.

Thank you. Thank you for having me.

Thank

you.