Webinar: How RPA might change the face of logistics
How RPA might change the face of logistics?
In this webinar, Shipamax Co-Founder and CEO, Jenna Brown, joins RPA Labs CEO, Matt Motsick, Let's Talk Supply Chain Host, Sarah Barnes Humphrey and Chain.io CEO, Brian Glick to explain how process automation creates cost efficiencies for shippers and service providers, and how it provides the backbone for future use of artificial intelligence.
ERIC [00:00:02] "Let's move directly into our next panel, I'm going to pass things over actually to Sarah Barnes Humphrey, host of Let's Talk Supply Chain podcast among a host of other things that she's doing."
ERIC [00:00:15] "So I really appreciate her taking some time today to join us in and oversee a panel that she was going to do at El Dorado again yesterday that was looking at how the what the impact of robotic process automation and just our mission in general, how that's sort of going to change the face of logistics in years to come. So I left. I'm going to turn things over to Sarah. Let her introduce her panel."
SARAH [00:00:41] "Thank you, Eric. I am so excited for this today and that we were still able to do our panel today. You guys can hear me OK. That's fine. Oh, good. All right. So thank you, Brian, for putting all of this together. I think this is, you know, just an amazing thing and we don't have a lot of time, so let's get right to it."
SARAH [00:01:04] "So our panel is the bots are coming. How RPA might change the face of logistics. So before we dive in, let's get to know the panellists today. Brian, you didn't get a lot of time to prepare, so I'm going to kind of throw a question in here, but I know that you've got this covered."
SARAH [00:01:24] "So let's start with who who you are, what you do, and besides your own tech. What is your favourite tech on the market? OK, Brian, I'll give you some time to take a look at. So, Matt, I'm going to start with you."
MATT [00:01:40] "My name is Matt Motsick, CEO of RPA Labs. And what we do is we provide software bots for the logistics and supply chain industry. The tech that I'm interested in are the robots that actually clean up the ocean. I love seeing that. And that's a good one. We can get a little bit more commercial on that and maybe get Merc to maybe add a cleanup on the back of their vessel. Yeah. You know, clean up some. You know, it was while we're well, we're going across the Pacific. Let's let's go in and clean up on the way."
SARAH [00:02:20] "Love it. That is a great one. Welcome Matt. Now, Jenna."
JENNA [00:02:26] "Hey, I'm Jenna, co-founder of Shipamax. We provide plug and play automation for text extraction. So, for example, taking unstructured data from bills of lading using machine learning models, and syncing that data in a structured way in your ERP or TMS."
JENNA [00:02:36] "Maybe keeping in the context of supply chain one of the coolest technologies I've seen comes cheating a bit because it's an integaration partner of ours, is a company called Sedna is an email client, but it's built for group email and it's programmable. And so you can really sync that up with your ERP systems and then do some pretty cool stuff."
SARAH [00:03:10] "Awesome. Love that. All right, Brian, you're up next. Everyone I mean, everybody knows who you are."
BRIAN [00:03:15] “I'm not going to reintroduce myself. I'll just so I'm looking at the list of people that are presented so far. And I think we four of our integration partners and we've already talked to five. I'm sorry, five of them, including this panel. So I'm going to avoid the industry a little bit, because I think so many. I'm honestly going to say Google Chrome Books."
BRIAN [00:03:37] "So I'm sitting on my desk right now. I have my Chrome book that I use every day and a Mac that I'm using for this session and the thing I love about the Chromebooks and this does tie back to the industry is they are truly plug and play. Right. You turn it on. Yeah. I know. If I drop it, I'm gonna be able to just pick up the next one. And you know, at some point when we look at all these standards and all this automation, it's really so when there's a massive shift in the industry like the virus or the tariffs, we should be able to do the same thing amongst our supply chain. Right. And right now, supply chains work a lot more like if you lose your Mac than they do, like if I lose my Chromebook."
SARAH [00:04:16] “Absolutely. I love it. I see. I knew you would nail that question even with no preparation. All right. So while I was doing some prep for this panel, I came across a quote from Gie Korten on LinkedIn and formerly of MFour. And I just want to throw this by you before we get into the panel. So unfortunately, I believe A.I. and machine learning have been hijacked by marketing efforts. We throw around the terms too easily, almost too lazy in our thinking. There has always been a need to infuse numbers and maths into strategy operations, supply chains, marketing, selling and all aspects of business.”
SARAH [00:04:53] "The power and reach of the maths has grown in large part to digital digitalization. End of the day we are trying to conduct. Business in a profitable manner. And so that leads me kind of into my next question. You know, tell the audience why RPA is not like AI and machine learning and why is it more practical and achievable to be more profitable? And I'm going to start with Matt on this one."
MATT [00:05:27] "I think that AI and machine learning are powerful. I think that. However, I think the words digital and digitisation have been oversold. And, you know, when a company says the word digital or A.I., it's almost like either they're putting something behind the scenes and they're doing things behind the scenes without someone knowing. A little bit with RPA, though, RPA is this really technology running scripts that you can actually see things happen. You can see, you know, the progress, you can see it working. So I think that's a little bit different, let's say, than A.I. and machine learning. But. Just just let's be clear. Robotic process automation requires A.I Machine learning to really make it happen. So it has to be used within the RPA room.”
SARAH [00:06:37] "Awesome. What about Jenna? Do you want to jump in here?"
JENNA [00:06:40] "Yes, I'll just maybe I'll just kick off some definitions to make sure everyone is on the same page. When I think of RPA I really think of back in the day when you're recording and macro and Excel and just runs through some actions and just repeats it. So I do actually think and you can have RPA with no intelligence at all baked in. Actually a lot of vendors in the market are in that position.”
JENNA [00:07:05] "I'm not a huge fan of the word A.I, but let's use machine learning, where machine learning can come in is any task which just requires a little bit more, let's say intelligence around a taxing role than just that kind of schedule based thing. And so I'd say there's a kind of new breed of automation softwares which includes machine learning. And and coming back to a question on how that can and making more profitable."
JENNA [00:07:38] "Let's say there are some highly repetitive tasks in logistics and let's say in our case like data extraction, very simply, if you can automate some of these tasks, it just helps you scale up more efficiently. So you need kind of less human bodies and doing those automated tasks for you to handle the same amounts of shipments."
SARAH [00:07:59] Absolutely. So why don't you like the word A.I.? Do you kind of agree with Gyi that it's over marketed?"
JENNA [00:08:08] "I think very few companies are using true AI."
SARAH [00:08:17] "Got it. Got it. Brian. To do and jump in here with any of that."
BRIAN [00:08:21] "Yeah. So while I agree that AI is a term without a definition, whenever you have that, it's a little scary."
BRIAN [00:08:29] "But what I would say about RPA, you know, I think I'm going back to sort of my role as a former CIO of a forwarder is there is technology that you can implement like sort of machine learning for analytics that you measure in months or years and you do a long project. And at the end you probably have something with RPA. It's like if you put in an RPA solution on Monday and you don't see that you're world's better on Tuesday, you can change it. But that's the time, not this or the time for ROI, but definitely the time for like, you know, if you're using Jenna's thing yesterday, you were typing in, you know, a bill of lading and today you're not typing in a bill of lading or, you know, that's not answering an email that yesterday you were answering by hand. So you very clearly know whether you're getting a benefit very, very quickly. It's much more precise than those other terms, which are sort of broad based and, you know, can be co-opted by marketing really easily."
SARAH [00:09:30] "Yeah. And I'm glad that you brought it up, because it puts it into real terms as to really the profit and the benefit of using RPA in your every day and instead of. And also when you're supposed to see not necessarily the ROI, but when you're gonna see that benefit, it really works in your favour. And like you said, you know, there's something wrong if you're not seeing it within a couple of days. You know what I mean? Like, if you're if it's a week down the road and you're not seeing the benefit. Then there's something wrong and you can deal with it right away."
BRIAN [00:10:03] "I guess, if we may have just set a very high bar from it."
SARAH [00:10:08] "That you guys want to weigh in on that before I move on from that challenge. Oh, yeah. All right. You heard it here, folks, on the panel at Virtual Eldorado. Jenna has taken on that challenge. All right. So next question. Actually, I'm going to start with you, Jenna."
SARAH [00:10:28] "We're talking a lot about talent in supply chain."
SARAH [00:10:30] "And I think there is, you know, a big gap between not being able to find talent and then talent being actually or jobs, I guess, being reduced because of RPA, especially in the 3PL world. So what are these? What are these realities? You know, 3PLs are very human dependent. So what exactly is RPA going to do for supply chain jobs?”
JENNA [00:10:54] "So I think there's a lot of kind of fear around this, and I don't think that needs to be said. I think as you kind of alluded and there is a shortage of talent. The industry is still growing and it is a skilled and skilled job. And that trend continues, automation will just help companies adapt to that new equilibrium."
JENNA [00:11:16] "And I also think for jobs it will make those jobs a nicer environment. So what we commonly see is that these are skilled people and a bunch of their time right now is spent doing really low admin, repetitive data entry and whatever tasks. You can take those away and start giving them the ability to add better value to their customers.”
JENNA [00:11:43] "And also, don't forget that there are new tasks that are coming on the market as a result of these technologies being in place. So, for example, and let's say more or more customers are looking for a better quality and quantity reporting something like that. Let's say right now, you might give your largest customers SKU level reports and you do not physically have the labour to provide that level of service to everyone. So automation, one allows you to kind of step up and provide that service to everyone. But then also humans are not having to do that task anymore and can come and add the extra analysis and service on top of that, to keep providing that customer service. And I think it's really no secret that the cost of acquiring a new customer is far greater than retaining your own ones. So, yeah, I I think it's I don't think we should be too scared about automation really impacting jobs over the whole."
SARAH [00:12:52] "OK. No, I like that. I mean, you know, like you said, there is some fear, you know? So how do we equate that to actually growth that's going to benefit everybody? And what does that mean? Right. Is it just about reskilling?"
SARAH [00:13:05] “I mean, that's a whole different panel and a whole different discussion. But so I think just because of time, we're going to go to the next one. Matt, I want to start with you. So what are some of the common use cases that you've seen so far on how logistics companies are using RPA?”
MATT [00:13:23] “Well, I think you got to look at also the issues, the current issues right now leading up to use cases. So, for instance, I'm just going to take logistics companies as an example. There's there's a couple of different problems that they're facing. One is they're receiving too many emails on a daily basis from customers, even though the logistics company says, go to my portal and get a rate.”
MATT [00:13:51] “The problem is all these other shippers are, you know, all these other forwarders and logistics comes, you're saying, go to my portal. So what do they do is they. They go back to what they do best and that's emailing out Plan B. You know, the free folders of logistics companies kind of get it right or where's my where's my shipment? So it seems to be you know, that's a little bit of the email overload.”
MATT [00:14:16] “Second, the second piece, really more of these logistics companies have cube farms, right? Of people just reeking and inputting those weightings, Robin notices commercial invoices in, you know, it's all about us. It's about, you know, basically reading those documents, you know, via RPA, reading, interpreting and then and then also just migrating that metadata, you know, into the TMX or the ERP systems.”
MATT [00:14:52] “So those are just some common cases. And then, you know, the other ones are just like workflow, just just automated triggered workflows. And these are just things that, you know, just common, you know, common, even tmw systems don't even have. Which is crazy because, you know, these larger TDMA systems, they have 60 screens, right? They only have six or seven screens if you RPA and automated processes are working. There's no need to double double key. And, you know, the same, the same different formats. You know, from the documents.”
MATT [00:15:32] “Just to give you a little example, you know, one things that we did was, you know, back to the, you know, the documents. Right. We have a free folder that does around two thousand shipments a month. And they have you know, they have so many people just reeking and inputting directly into the system. So what we're doing is we're basically automating their whole process, going from, you know, you know, someone going into the document and retrieving everything in. Instead, what we're doing is our RPA is automatically seeing opening up the attachment from the email, opening up that document, identifying and classifying what type of documents it is. And then and then taking that metadata and pushing it into systems.”
SARAH [00:16:25] "So I'm I'm I'm sorry.”
[00:16:27] "I'm just gonna jump in here with that. We've got a question from the audience. And I think it's it's kind of timely on this one. So how do we evaluate logistics partners and customers capabilities? When you engage with someone, how can you know that you're up against what you're up against before you dive in and commit to a plan? And that question came from Stephen War. Do you want to just answer that since you were kind of talking about the examples?"
[00:16:53] "Yet to me, I think I think shippers are now evaluating logistics providers on response time. Quite honestly, if after taking too long to respond to a quote to basically get shipping documents out back to the customer, you're taking way too long. So, you know, back in the 90s and 2000s, you know, maybe twenty four hours was fine. But no, it's it's been in the minutes to hours for us. So awesome. I think that's how RPO can really speed up how how companies react to and respond to their customers.
SARAH [00:17:32] "Jenna, I know you want to jump in here on this one."
JENNA [00:17:35] “Yes. And I think coming back to like how I think the question on how customers can evaluate all these different vendors. And I think there's two things to us. I think one and I really encourage people to go back to the problem they're trying to solve and really detail what that is, because I do see a lot of time when people go out to procure a software, people for all these kind of features out there, largely all great, but not valuable when solving that problem. So that will just keep us really clear. And then second to that, most companies, especially if you're thinking about that, the place piece like data extraction, we'll be able to run a small pilot to show you the end quality of that before you have to go and commit to those kind of annual contracts.”
SARAH [00:18:24] "Right. And then I guess on that same note. Are there misconceptions that need to be overcome when it comes to RPA? Like what are some of the objections that you guys are coming across?"
JENNA [00:18:34] "The biggest thing I have seen is companies seem to know that they have a problem with this. Picking the right tool for the job. I've usually seen people be burned by this before and from one of two things. Either, one they're just an RPA to do kind of data extraction or OCR and that's too rule or they're using a very generalist product, which is kind of promised the earth, but actually in reality is not suited for the task. And so I think that that's the biggest thing we've seen, just failed projects. So yeah, I think that comes back to that. And I guess really I think that the technology's given pilots and see the data extraction, and you should be able to see that pretty quickly with technology as well."
SARAH [00:19:24] "And what about not using the solution to the fullest?" I think that that's a really big issue. And I think Matt wants to jump in on that one."
MATT [00:19:34] "Absolutely. So what I've seen is actually, you know, there's actually been articles on adoption rates of RPA. And I think it's coming from the companies that just sell the frameworks only to the companies. And so when the companies say, great, I just purchased an RPA framework, the framework company or the company doesn't say, oh, well, you know, they do it after the fact, but they say, no, you've got to hire design developers, you've got consultants to implement that framework in order for your business. And I think that's something that Jenna and my company are doing, is that we're actually building, you know, basically full solutions to where they don't have to. They don't have to build out a framework. They don't have to hire developers to do that. It's pretty much already built in and ready to go."
BRIAN [00:20:31] "Can I jump in on that one? Yeah. Just so yeah. Is that from an integration standpoint, we sort of have that same approach that, you know, go by VDI tool, but it doesn't have all the logistics knowledge built into it and you have to still hire to that knowledge. And I think what a lot of industry specific tools do, whether RPA and integration it's the same, is boring. What are you supposed to do? What are the best practices? Right. How you know anybody, theoretically, you could drop a couple million dollars into a machine learning team to recreate what Jenna has. But why the hell would you buy it? And I think that that domain expertise is really the thing that if you're evaluating a partner. Same as hiring somebody to move your freight. You know, anybody can be a freight forwarder, but not everybody can get your shipment to move during peak. Right. At that level of expertise. I think it's huge."
SARAH [00:21:27] "Great. Well, so we've got we've actually got quite a few questions. And I know that we've only got probably about 10 minutes or so. So I'm gonna get to some of the audience questions right now. So Mike Cadieux says, I'm talking to a lot of RPA solution providers that have prospects coming out of the woodwork, but they think RPA is a cure all. How many engagements are you seeing where the client process really isn't a candidate for RPA because it is in constant flux versus processes that are well-suited for RPA augmentation? Do you want to take that one, Brian?"
BRIAN [00:22:01] "Can I pass on that one real? Yeah, sure. So that is not a bad thing."
JENNA [00:22:08] "I can. I can answer quickly."
JENNA [00:22:10] “I mean, we and I think it's really understanding. I'm going back to that problem on what it is you're trying to solve. Our company we only handle and data extraction really. And then when you need to go and automate tasks within systems, we say actually now you should look for a vendor for that."
SARAH [00:22:33] "Yeah. So it's really, you know, up to the customer to really identify what those issues are first and sort of work from there. Right. Because some solutions are going to work for some of the tasks that you're doing and some are going to work for those tasks once they are automated for you. Right. Further along the chain, Matt.”
MATT [00:22:53] "That's exactly how we come into companies, we say. Where's your biggest leakage in a month basis? You know, where are your biggest inefficiencies? You know, what's what's causing a little bit of your pain point.".
MATT [00:23:08] “And so, you know, that's really what we do first, where we're a little bit different. I think in general, I think we're just more of maybe like an RPA facilitator to where we're actually doing a combination of conversation, email, workflow and documents.".
MATT [00:23:24] "So we're really kind of really along that whole the whole path of RPA. But yeah, you're right. You have to know what you want. What probably won't solve first."
SARAH [00:23:41] "Right. I mean, there's a lot of challenges. We've got a lot of different areas that we need to take a look at. And it's really about identifying what those pressing ones are first and then, you know, starting to look at the tools that can help with that. And then again, down the line. So Nick Chubb says, do any panel members have a sense of the financial cost of manual processes and errors today and what might what that might look like with the high adoption of RPA, Matt?".
MATT [00:24:12] "Out of four hundred and eighteen billion dollar. Wow, that's operational costs. That's the ocean. That's basically all of transportation. So that's how big the market is. Huge. Yeah. And so if you can, you know, at least take 5 percent of that out of the market, you know, 10 percent of the market. I mean, that's that's that's really huge.".
BRIAN [00:24:39] "Think about it. Think about this. It's always the sort of way I think about it when somebody is double king and swivel heading. Between two systems. Yes. That time that they do that. I don't even care, to be honest. That's not the cost. The cost is when they miss the key, that arrival date. And then there's an email sent from the shipper that has the software provider on it. It's got that forwarder on it. It's got that 13 vendors on it. And it's this and that. And then you have to go attend a data quality session and you have to bring your executives that money, the money of the explosion, of the emails and the cleaning up the messes. Yeah. Me as when I from an operational standpoint, that was what we always worried about, right. It was, you know, a hundred emails that came from the one mistake that was so much worse than the one employee's time to type debated."
SARAH [00:25:32] "Not including the phone calls."
BRIAN [00:25:34] "Yeah, exactly. So. So you take all that right to not make the mistake because the robot is going to be more consistent. It is priceless.”
MATT [00:25:42] "You can look at any technology out there in the industry. But RPA and integrations are going to change the way. Logistics companies work today. Pretty soon, like in five years, you know, you'll be able to, you know, as an operations person in a logistics company. You'll be able to say, do I want to do this or do I want a bot to do this? And then you can just relegate whatever the manual task is, you know, to the boss to actually, you know, do some other things. So I think there are some that are going to definitely change the way logistics companies work.".
SARAH [00:26:22] "Well, and I think that, you know, it's taking it from a reactive environment into more of a proactive predictive environment is where we all need to get to anyways, because we are in a chain. Right. Supply chain. Right. Everything that we do affects didn't. Different parts of that chain. And you're right, Brian. If we make one mistake in one area, it explodes and there's a huge mess to clean up. And with the technologies and different things that we have out there, we shouldn't have to do that anymore. I mean, eventually kind of wishful thinking, I guess, at this point, but we're getting there." So now Leisz says companies that have in-house RPA teams, are they. Are you guys open to working with them for specific projects or do they tend to complicate the process? Good question."
BRIAN [00:27:17] So do you guys want me to take his chances on that song in our future? So they're all going to have to go talk to those teams, because I can tell you what it's like from an integration standpoint when we go into a company. Because most forwarders have some sort of VDI or integration in-house already. And what we find is that in the early days, there is a lot of change in resistance. There's a lot of, hey, you know, we're here to do this. Are you telling us we're not good at our job? Is there, you know? Are you just outsourcing our function? And what we work very hard to do at the beginning is establish that, no, we're here to supplement and augment. And maybe you're not. Maybe you won't be quite as hands on the keyboard. But there's still a lot of institutional knowledge and depth that's important. And what we find is that it's our job as software vendors to make those people into allies where, you know, I like to say to my team that after the first kubey is if the executive says they're going to fire us, that person should threaten to quit. That's how you know that you're providing value to the team as well. So, yeah, there's definitely change management and definitely resistance when there's an in-house team.”
SARAH [00:28:31]” That's you know, that's a great answer because. And change management is coming up in every single conversation that we're having. You know, it's the leadership skill that everybody needs to have because at the end of the day, everything's changing so fast, whether it's RPK machine learning. And that is really going to be a key skill to be able to navigate those waters and figure out what's good for your company. Figure out what's good for your teams and how we can all collaborate and sort of make that happen.”
SARAH [00:29:01] “So I don't have a name on this one, but this is a really good point. So this person says different logistics companies may use different technologies. And, you know, being working at a 3L for as long as I did, I know this to be totally true. You know, they've got different systems, W-M system, CMF systems, you know, accounting systems, that kind of thing. So he's saying, can automation solve this problem to bring everyone under the same umbrella so that users do not need to go from different platforms to choose different carriers or I guess, you know, different functionalities within their companies.”
MATT [00:29:40] “Well, I think Brian has a really good product for this. You know, with the integration that he has, the integration solution that he has. I can say that wherever there's not an API, then that's kind of where, you know, RPA kind of comes in, is using those software bots to integrate between systems that are not able to use a desire or or API. So I kind of consider RPA is API on steroids, quite honestly, with how you can kind of connect systems with that thing.”
SARAH [00:30:14] “Yeah, that's a really great analogy. Jenna, did you want to jump in on that one quickly?”
JENNA [00:30:19] "Yeah, I mean I yeah, I do think and you know whether its RPA or a kind of integration is absolutely, all these automation technologies can sync up the systems.".
SARAH [00:30:31] "Awesome. One last question from Amir that I just saw, RPA versus orchestration. Does anybody want to jump on that one?" Is there a difference?”
BRIAN [00:30:43] “Yeah, I mean, so orchestration. Think about the whole swim lane RPA could be dealing with an entire business process, but often is dealing with a piece of the business process. So you think about orchestration might be everything. Coordinating to know whether it's from origin all the way to delivery or whether it's just origin onto the ship that could have an orchestration. And then you may use RPA and integration and humans and physical robots all within that to solve different pieces of that orchestra."
SARAH [00:31:15] “Yeah, and that's a really important piece, right? We have to remember that it's all of those different pieces that really are going to collaborate and move everything forward. OK, so before we wind this down, I don't want to start the conversation, but I kind of have to. Time is dictating. I'm going to let all of you guys answer what is one task that everyone in this audience needs to look at in their own supply chains that could be automated right away. That would bring efficiency and cost saving opportunities.”.
SARAH [00:31:46] “Jenna, go.”
JENNA [00:31:48] “Manual data entry from documents.”
SARAH [00:31:51] “All right. Matt.”
MATT [00:31:55] “Customer response, because if you can respond back in the customer first and B, the early bird gets the worm, then you will get the business.”
SARAH [00:32:04] "And customer experience everything these days, right? It dictates everything down the chain.".
BRIAN [00:32:12] "Brian, great payment. It's boring. It's not customer facing. And it's a great place to experiment because it's a pure cost centre where you save money. Interacting grew up your first project and its freight payment. It's only your vendors, not Europe, not your customers who are upset. So that's why I like to start there.”
SARAH [00:32:32] "Good point. Good point. All right. Well, thank you, everyone, for the panel today. I think it was a great discussion. Hopefully the audience got a lot out of this, too. And then I guess it's back to Eric."