The chargeback you don’t see until six months later

Episode 16 - The chargeback you don’t see until six months later - Transcript

Matt: Welcome to Data Versus Commerce, where we explore the messy middle between database and doorstep. I’m Matt Johnson.

Floyd: And I’m Floyd Blaikie. Let’s dig in. Okay, welcome Leo — you’re brand new to the podcast, we’re so excited to have you. Can you start by telling us a little about yourself? What do you do, who do you do it for, how long have you been doing it?

Leo: Yeah, great, pleasure to be on this podcast. My name is Leo Rodriguez, I’m the VP here at River Plate, Inc. We’re a 3PL in Southern California, been in business for 34 years now — I personally have been here 17 years, going into my 18th. Retail, distribution, e-commerce fulfillment, wholesale distribution and fulfillment, value-added services, cross-docking, we do some freight logistics too — a little bit of everything, pretty much multi-channel, omni-channel fulfillment and logistics solutions. That’s who we are — we have brands in a bunch of different verticals, a lot of different commodities we’re servicing, and it’s been a fun ride.

Floyd: Oh, 17 years in fulfillment — this could be the whole episode. What’s changed in 17 years in fulfillment? Maybe fast-forward a bit, then really focus on recently, because I feel like the rest of the digital supply chain has really changed. What’s been going on in your world?

Leo: Look, the origin — it was a family-owned business, my mother started it doing trucking and freight services locally in Los Angeles, and that turned into just listening to clients and what they needed — it wasn’t just picking up freight and pallets from point A to point B. They needed a pit stop for a month or two or three, whatever it might be, so that’s when we looked into getting a warehouse — this is 30-plus years ago, before I was involved. Warehousing services then turned into needing to label and create kits, and here we are now, at a 138,000-square-foot facility in Simi Valley, California. So we’ve definitely grown a lot, learned a lot, burned to learn, to say the least — but hey, we’re still here, we’re thriving.

Floyd: Okay Matt, we’re going to play a little game — I’m going to read you a post from Leo’s LinkedIn, and I want you to tell me what you hear in this. I’ll paraphrase a little so it’s not story time, but he starts out: “In fulfillment, what happens on the floor has to match what’s in the system every time.” Then he’s got a little illustration, we’ll get into that maybe later in the show — but after that he says, “Physical and virtual have to run in tandem. When they don’t, that’s where chargebacks, inventory discrepancies, and compliance issues start. The operation is only as clean as the data behind it.” Does that sound familiar to you, Matt?

Matt: Oh boy, yes. It’s the reality of the world we live in, and it’s really what differentiates the Amazon of the world from everybody else who thinks, “Wow, this is going to be so easy to sell online.” But finding the product is only the beginning, buying the product is only the beginning — it’s about getting the product and keeping the product. The getting and the keeping is where supply chain matters, where data governance, data accuracy, system integration make that work. It’s something we see a lot, Leo, and I’m glad you’re solving for it. I’m curious — not to take us on too much of a tangent, but I’m a family business guy myself, so I’m curious about the digital transformation you saw. I’m guessing you and I are probably about the same age, so we probably have a similar story — what happened when you first came into the business, what does it look like today, and how have your customers’ demands driven some of the innovation you’ve gone through?

Leo: Gosh, when I first came in, it was learning the ropes — I didn’t have a supply chain or logistics background, I actually went back and took a bunch of courses online and really tried to educate myself, spent a lot of time in the operations on the floor picking the brains of some of the veterans at the business, and trial and error, of course. I was doing my own pick and pack and being an account manager, communicating directly with the different brands, but I noticed immediately there was a big disconnect — they’re taking in orders, and there’s a lot of excess fat, you have to find something to better streamline it. The spreadsheets can only get you so far, a lot of manual entries that turn into errors. We were like, “Okay, hold on — these shopping carts online need to get those orders into our system to allocate to inventory, you need specific rule sets to establish the distribution or fulfillment protocol.” I just went, out of pain points, into looking at what solutions were out there — cloud-based systems, warehouse management systems, that was the first thing we wanted to tackle. We tried one at first, a smaller company — it was great to learn the ropes, but we immediately pivoted into a larger company, they’ve grown quite a bit since too, we’ve been with them a very long time. A lot of it is understanding there’s so much within these configurations and settings — not just the company profile settings you need to set yourself up for success on the operations side and for the account team to execute their work, but the alignment that has to run in parallel with the client customer profiles — their business rules, their trigger points and notifications, how their data streams into yours, essentially push and pull. And then, probably something you see quite a bit and we tackle all the time, is trying to decipher what’s bad data and what’s good data.

Floyd: Leo, you said something to me before we started recording, and I’m eternally mad about it because I wish I’d captured it the first time — you said data is a chain reaction. I love that framing, we talk about that all the time, how data decisions and governance, and how much you invest in it upstream, really dictate how that whole digital supply chain is going to flow. We’ve talked to OMS experts, we’ve talked to e-commerce experts — you’re like the last line of defense, though. We haven’t talked to a Leo yet, and frankly I want to hear some horror stories. What’s happening when you’ve got bad data, or incomplete data, or data that doesn’t match reality upstream? What do you see on the warehouse floor as a result of that?

Leo: Yeah, it can go in a lot of different paths. If you’re not collaborating between the brand and the 3PL — not just in onboarding, but going deeper, spending an hour on your part of the research and development, what you’re doing on the manufacturing end, what your goals are, whether you’re doing e-commerce and wholesale and retail business, are you sharing these SKU profiles, are the inner packs and master packs, as they’re discussed with your supplier, actually set up for success — there are so many variables that need to be taken into consideration. But to answer your question, the horror stories are that even with some big brands, decisions get made without diving in deeper, without having those calls and meetings with logistics professionals to go through some testing, talk about scenarios where things could jam up. Because a lot of times, since we’re the last ones to touch it before it hits a DC or warehouse floor and then the customer or recipient, that inventory they projected — “Okay, we’re for sure going to get this within eight weeks” — and if things happen in the world where that eight weeks turns into 12 weeks, and now they have extensions, do they really have that much time for reworks? You’re just adding a lot of extra layers of pain — maybe cut POs, loss of sales. So we’re here a lot of times to review and help them protect those margins.

Floyd: How does that show up in the data too? You mentioned things like changing the number of units that ship per box, or changing the packaging in some way — that’s often thought of as a physical change in logistics, but how does it impact the data you’ve been sending along with that physical product throughout the supply chain?

Leo: Sure. For example, we’ve seen item profiles set up months and months in advance, pitched to retailers or specific marketplaces — this SKU or item profile is going to be a 12-pack, say, and they set it up that way, their supplier portals and EDI mirror that information, making sure it’s set up for success. Then two or three months later they realize, “Oh, we’re also going to be doing this for this channel or marketplace,” and now they want it in a six-pack or eight-pack — but they’ve forgotten they already set it up as a 12-pack, and that’s what they sent into our system, that’s what we were also set up for. We verify and check dimensions, weight, and unit measures, but sometimes we’re not privy to the supplier admin setups on their back ends. So we’ll see the POs come in, we’re ready to go, and then, wait — hold on, these unit measures and the packaging unit measures the PO is asking for don’t match. The data virtually is lagging, it’s behind, it’s not correct, and it’s not an overnight quick fix — some of these systems, you’ve got to go through the retailer channels to revise POs. So a lot of times, for a 3PL, it’s tough, because the back-end upstream data affects the downstream, and we’re not always privy to all those portals and setups. And nowadays, with a lot of these marketplaces and retailers, with OTIF and all these performance metrics, you’re on a report card — and unfortunately a lot of that turns into chargebacks you could have prevented. So accurate data is utmost critical in terms of passing through our systems from their EDIs or OMSs down into our WMS, into our TMS, and then pushing back accurately.

Matt: I had a follow-up on that — specifically, as a good 3PL, you’re doing a lot of consulting work, I’m assuming, on the front end and throughout the relationship with the brand, so there’s a continual feedback loop, and data quality is probably a regular part of that conversation. What have you guys been doing, or what have you seen in terms of best practices, for facilitating the QA of the data as well as that feedback loop with the brands?

Leo: What we use the AI tool for is more to point out anomalies — not a large data set, just what doesn’t match in terms of what we’re prompting it to do. There’s quite a bit of that right now, and we’re still learning and educating ourselves. With our WMS, we’re just about to jump into an AI feature where we can upload SOPs for both the client-facing and our own operations, with separate user permissions and rule sets. Our clients can, even for something as simple as onboarding — everything we upload becomes essentially a library, and they can query it based on exactly what we did during onboarding, how these retailers or marketplaces operate, how they label things, and just prompt it, and it’ll pull from that. Same thing from our operations side, and it can do so in different languages — they can ask, “How do I label inner packs for Kohl’s?” and now everyone’s universally looking at the same information, able to query these specific documents, rules, or workflows in a more organized way.

Floyd: So all the major retailers, most people doing retail or e-commerce in general, are familiar with EDI — very structured, very clear expectations, here’s what you’re going to get from the retailer, here’s what you’re going to get back, your purchase order, your acknowledgement of the purchase order. It’s all very structured, and that helps avoid the manual data entry errors that cause bad things to happen on your floor, Leo. On the other side of that, the people who aren’t in the warehouse, the people in the boardrooms — the “big brains,” air quotes for anyone who can’t see — they’re talking about APIs. I think there’s this utopian future where data flows between all these different AI platforms, order management platforms, and warehouse management platforms via API, you’re querying clean data, getting exactly what you need, and humans are kind of hands-off. I think I know the answer, but how far from your reality is that? Is that something you can ever see happening, or is that something you just laugh about while having your 5 o’clock beer in the driveway?

Leo: Well, look, I hope so — there’s been some advancement with some of our tech partners in terms of opening up the API source to connect to more fields, the REST API specifically. Right now we have a developer’s link we send to clients who aren’t doing a one-to-one, pre-mapped integration like a Walmart Marketplace or Shopify — those are just push, put in the API keys, it connects, but you’re put in a box in terms of what you can and can’t communicate. Hopefully, more and more, we’re able to be a little more virtually nimble in terms of the data we need, more client-specific, so we don’t roll into exception handling. But look — we just had a client, for example, a doctor, actually my nutrition doctor. He’s doing a refurbishment program, reverse logistics specifically, and he said, “Look, I’m just a one-man show here, I want to be able to connect” — he created his own dashboard and operating system to manage this reverse logistics via Claude. It worked great, and he was doing it from his clinic. All of a sudden he got to like 120 shipments back to him every month, climbing toward 200, and he said, “I don’t want to be doing this anymore.” I said I could help him out — sent him the mapping, the developer’s link for the REST API. He plugged that in, prompted it in plain English — he doesn’t understand coding — into Claude. Probably the fastest connection we’ve had, processing through our returns feature in our WMS, with everything in the right field, accurately, based on our requirements. He said he did it in less than 15 minutes.

Floyd: This is the second big win for Claude I’ve heard on this very podcast in the past two or three episodes — Matt, you were saying Claude did better than some purpose-built AI product data enrichment tools. What’s going on over in Claude land? Is Claude the future of commerce? Let’s talk about it.

Matt: All I know is they need to reach out to us and set up a sponsorship — we can put together a nice package for you guys.

Floyd: Salesforce did it, why can’t we?

Matt: No, it is really cool. What it goes back to, and makes me think, is that when you have business-minded people who understand the customer experience they want, how they want to work with their partners and software, they just never had the ability to take action on that vision before. That’s what we’re seeing — smart business people being able to do technical things they’ve never been able to do. In the 3PL and reverse logistics world, it’s a perfect example — I know what I want it to look like, I just don’t know how to make it, and boom, now Claude’s here, 15 minutes. I don’t know if I believe that, but —

Floyd: Claude and Leo.

Leo: Yeah, that’s what he — when he told me, “I did it within 15 minutes,” I was like, “15 minutes once you were ready to connect it, I’m sure.” But the fact is, I gave him the link, and the next day — he’s a busy doctor — he already had the test returns in there, a bunch of them, and sent me over what Claude had prompted, saying, “Does this match up with everything you required from him?” I said, “Yeah, everything lines up.” There was a lot of customization in what he wanted to match — like you said, it’s, “This is what I envisioned in the customer experience,” but also taking away a lot of the inefficiencies or excess monitoring — he could get all of this into his system and ours, so we could go ahead and do our quality assurance program, quality control procedures and services. It was really cool to see.

Floyd: So for some of your larger customers — no offense to your doctor, that does sound very cool, but —

Leo: Yeah, that was just — I wanted to help him out.

Floyd: No, that’s awesome. But is there a red flag for you — if you’re onboarding a new customer, is there something that happens in the first 30 days that tells you these people don’t have their product data together? Are there immediate problems you can see, or do they tend not to show up until later in the engagement?

Leo: No, it’s not that difficult to point out, and we’ve seen it multiple times — you have established businesses, 50-plus employees, a lot of different departments, procurement, fulfillment logistics, an international logistics team — and there’s disconnect between all of them, they’re not passing the correct information internally. Unfortunately, sometimes you see team members on the brand side who want to do the right thing, they just don’t know how — they’re not as knowledgeable about how the supply chain works, or how to work with a 3PL, and they struggle. We try to help them out, but then they’re providing not the correct information, and we have to lean on them to provide us with the correct data. Sometimes, even with big accounts, they’ll take three or four days, or not pay attention at all — you’ll get five bullet points, they answer two of them, and brush off the rest with, “Oh, I’ll get to that later, it’s not that important right now.” No, it’s very important right now — this is probably the most important time to get the setups and configuration right, otherwise everything else snowballs, and it can create a lot of problems. Some of the tighter-knit teams, though, it’s beautiful to see — they’ll take any opportunity to pick our brains, get on calls, collaborate, communicate. You save a ton just from one hour of operations calls, or integration setups, or pivoting on a parcel program — down the road you’re saving thousands, tens of thousands of dollars, just from having those conversations instead of, “Oh, it’s connected, it works.” No — connecting doesn’t mean it works. You need to connect, test, and even go past go-live to actually prove it’s working all the way through the loop.

Matt: And maybe — correct me if I’m wrong — it’s also about making sure there’s a very clear, documented process for when something changes, because I think that’s probably where most errors happen. Engineering makes a change, or marketing makes a change, packaging changes, and it never actually makes it through to the supply chain data — if that’s not clearly defined, and we don’t have all those scenarios mapped out for what happens when it happens, I’m guessing that’s also a failure point.

Leo: Yeah, you nailed it — it’s actually common. We try, between the brand and our team, to stamp it into their brains: please, any changes in item profile setups — even if it feels redundant, like, “You guys work with Walmart with five, seven brands, why do I have to send over the new compliance guideline again?” It’s because we capture and archive that in your client profile, and now, more so, as we’re putting that into the AI, we make sure it wipes out the old one and puts the new ones in — because there might be something like, “Wait, hold up, they want us to route a little differently here.” Just an example. It happens quite often, but you learn from it — our account managers and CSMs communicate that with the brands, and hopefully they create a fix. It’s constantly evolving, constantly changing, for domestic and international business, and you have to always understand you need to revisit all your marketplaces and profiles every couple of weeks — not necessarily every single item profile, but something within your operations or company profile, every couple of weeks.

Floyd: That’s wild — it seems like the kind of thing you’d have nightmares about, because you could have your product data perfect, immaculate, correct — somebody found the thing they’re selling, eight to a box, the SKU was complete, all the attributes filled out, they’re getting exactly what they want — but you missed data about how it gets to them. Suddenly it’s not getting to them anymore, because even though they’re only going to see one of those eight boxes shipped together, you forgot to tell your logistics provider they were coming eight in a box instead of 10 in a box. All of a sudden there’s a problem. So it’s all these different dimensions of product data that need to be absolutely perfect to make the supply chain work. Not a question, I’m just reliving some of my own experiences.

Leo: Yeah, no, these are facts. And look, a lot of what I’m talking about is “the brand this, the brand that,” but there’s also a solid percentage of it that’s out of their hands. Something we’ve had recently — mapping changes on the back end of a tech provider that screwed up the connection, and you’re trying to figure out what happened in that specific integration, reaching out and hearing, “There’s something going on” — back to EDI, there were mapping changes we didn’t see until ASNs were sent out, and all the way downstream it wasn’t visible, because you get alerts and warnings, but if that warning isn’t communicated either, then there’s an issue. All of a sudden you get a random email from someone at Target saying, “The last two weeks, all these ASNs have been coming in incorrect,” but on our end it’s showing a green checkmark, data looks fine. No — all the way down to the retailer EDI, something got out of whack, and then it comes back, “Yeah, there were a couple of changes done internally.” It’s not about pointing fingers, it’s about fixing it as early as possible, because otherwise you’re spending all this time fighting chargebacks you don’t even see until six months down the road, when they’re actually billing for it — might be exaggerating on six months, but a few years ago that was really the timeline for some of these big-box retailers, holding onto those for a long time before saying, “We could have had corrective action a lot earlier.” They’ve gotten better, but a lot of times it’s just not seen until it’s pointed out by our clients’ clients or partners.

Matt: I feel like we could have done multiple episodes on all the different players involved — the software, the retailers, the distributors, it’s all interconnected, and the supply chain is like the ultimate team sport, it takes a group effort to take something from a digital product into the hands of the consumer, the buyer. If a good 3PL is sort of like a coach, what would you coach our listeners on? What are some best practices, some thoughts you want to leave with the audience, that they can take back to move the needle a little in their operation?

Leo: Good question. If I could give a brand one piece of advice, I’d say: three weeks before onboarding, test the reality, not the PowerPoint or the boardroom decisions. Really understand what you see in terms of your shipping rules — as a consumer, what do you see as a B-level experience versus an A-level experience? If it’s an A-level experience, what do you see in terms of the deliverables on the surface, once you get to it? But ask your partners how they think they got there — from your experience with other brands, what was the prerequisite work they did, and try your best to mimic that. You have to be relentless about it. It can be boring, but I’ll repeat it again — test beyond connectivity, put it in a test environment, do pilot runs and sandbox environments, really put the work in to set yourself up for success. Spend that critical time in the beginning, and then downstream you’re not sitting there with four or five different channels still fixing things that could have been fixed during onboarding, battling because the brands are jumping all over the place. That can become a big issue. We see the difference between the brands and clients who do that versus the ones who don’t — they’re operationally a lot smoother. Over the years I’ve been here, I’ve seen the ones who make that effort scale their business in a way that’s been impressive and successful.

Floyd: Thank you so much, Leo, it’s been awesome having you on the show. Before we let you go, if people want to know more about you or River Plate, where should they go?

Leo: You can go to our website, riverplateinc.com — we have an RFQ page you can fill out, kind of a small questionnaire with our sales email. A lot of times we’ll do a discovery call to see what your goals are, see if it’s a right fit — because in our world, we don’t want a brand relationship that’s short-term and doesn’t work for anyone. This is a big investment on both sides, and we want to make sure we set everyone up for success. We’re not the right fit for everyone, and there are plenty of other awesome 3PLs out there, so definitely do your due diligence and homework. But you can reach us via LinkedIn, our company profile, or even make a phone call — we have team members who’ll pick up the phone and set up a scheduled call with someone from our sales or business development team.

Floyd: All right, thanks so much, Leo.

Leo: Appreciate it, thank you for having me on, great conversation.

Matt: Thanks, Leo. Thanks for tuning in to this episode of Data Versus Commerce. New episodes drop weekly. So if you’re responsible for any part of how products get from a database to a doorstep, subscribe now on Apple, Spotify, or wherever you listen.