How Sifted Helps Shippers Monitor Parcel Spend Before It Spikes

The real answer is to proactively track what's going on within your shipping and notify you before you know you've spent a lot of money. If you've got to go back and save and collect money, that's money that you spent. We want to uh project what you're doing and prevent that from happening uh in the get-go. And so that's really the impetus behind the evolution of our product.

And you know, I say we started out as a managed service, we transformed to a SaaS business, and we're currently in the process of transforming to an AI business. Welcome into another episode of our AI Use Cases in Logistics series from Cargo Rex. Today we are talking with Sean McCarrick. He is the CEO of Sifted.

It's a SaaS-based business optimization and parcel spend management platform. So, Sean, welcome in. I'm excited to have this conversation. Thank you.

Great talking to you. Now, for folks who may not be aware of your platform and sort of your ICP, your ideal customer, give us sort of that high-level overview of where Sifted came from and then who are your target customers. You know, Sifted emerged out of the consulting and audit business that goes back 20 years and parcel shipping companies coming in and consulting, helping shippers figure out how to uh reduce their cost and auditing invoices. And it's a very reactive, backward looking.

Hey, you spent a bunch of money that you need to spend, I could save you like $500,000, uh, etc. And VeraShip, when I started, I think was 15 years old, and we went down the path of really transforming it to a uh a SaaS business and really uh kind of completed that journey uh in 2024 and basically building a product that shippers can use themselves to proactively manage their shipping spend. Uh um and then we most recently we attempted in 2023 actually to uh build an AI uh client facing AI solution technology really was not there yet.

And so most recently releasing our co-pilot um product, if you think about the evolution of SaaS, SaaS has evolved to being a very efficient, you attempt to build a very efficient tool to make it easier for people to do work. And really with AI and our co-pilot solution, we're moving from a place for our shippers to do work to a solution that does work for them. And so um, you know, the intent is you don't need to log in, you don't need to like come visit the uh our product, you are set up alerts, you set up triggers, and we're monitoring your shipping spend.

Uh um, and then we communicate proactively to you when there's a variance that you need to take uh uh pay attention to. And the big change in this industry, you hear is savings a lot. And in reality, if you have a big ticket number that you can save, that's a problem. That means you already spent the money.

And so the real answer is to proactively track what's going on within your shipping and notify you before you know you've spent a lot of money, right? So we should avoid like needing to like if you've got to go back and save and collect money, that's money that you spent. We want to we want to uh project what you're doing and prevent that from happening uh in the get-go. And so that's really the impetus behind the evolution of our product.

And you know, I say we started out as a managed service, we transformed to a SaaS business, and we're currently in the process of transforming to an AI business. And so, what do you think you've learned from those two initiatives that have uh kind of put you in a unique position to where you can say these are the red flags or the early warning signs that you should be on the lookout for? Is it you know years of data, or is it maybe the years of data plus you know, the the layer of the customer-specific data?

What are maybe the great question? Uh um because my criticism of our, I'm gonna call it legacy solution now, is that hey, it's a wonderful tool. It has uh um, it's very complex, it covers a lot of ground. Uh however, you have to know you know what you're looking at.

So we have what 20 years of experience in partial shipping. We have built that into our software. Uh however, like I said, you you needed to know what you're doing and uh invest the time and energy as opposed to having a natural language interface that says, hey, uh really, I want to know if my cost per pack increases within a variance of one to two percent. I want to know the following things.

Like, so you don't have to be now you don't have to be an expert. You're gonna talk to uh our natural language interface like you know, it's uh an employee for you. And that's really the way we position the product. This is an employee for you that's gonna do the work and you're gonna interact with the same way you would, same way you would one of your employees, but simple.

And so on on your website, I love how it's laid out with all of the different use cases because that makes it hyper specific. And so for a lot of these, you know, you have finance pros, logistics pros, uh, for 3PLs, uh, for manufacturing, all of those different use cases. Is that at the enterprise level, or do you have some some SB type customers, or is it maybe a blend of the two? Uh I would say uh um mid-market in the enterprise.

You need to have enough parcel spend that it is worth your time or you it's necessary for you to manage it. And it to be candid with you, it really, you know, parcel spend is something that's kind of uh snuck up on the industry and really exploded post-COVID. Uh and so it really comes down to the ICP, really is uh comes down to the um sophistication and maturity of that particular shipper and desire to get their hands around uh what's impacting their gross margin or what's impacting their PL.

And so uh we have shippers that are in the high SMB mid-market all the way up uh through large enterprise that want to control their shipping spend and understand what's driving it. Outside of, hey, I'm gonna have somebody come in and do some consulting with me or uh audit some invoices. No, that that that definitely makes a lot of sense because especially since uh everybody on their mother, well, I don't want to say everybody on their mother now knows about supply chain, but you know, ever since you know the Evergiven got stuck in the Suez Canal, you know, the supply chain managers and VPs and executives seem to have that seat at the conference room table now or at the boardroom table.

And so they want to have their their lives, you know, kind of made a little bit easier with with some different you know products that can give them that eagle-eye view and help to explain you know cost savings, which everybody understands and everybody wants. And so, how what does I guess sort of the the one? Cost avoidance. Like I, you know, I would tell you that on average our clients see a 17 to 20 percent reduction in cost per pack.

The more important part of that is avoidance up front. The way you get the number lower is by making sure you don't incur things to begin with. And so, and you know, you were you were going through those profiles. What we see is that depending on the company, partial spend might be managed by operations, it might be by supply chain, it might be managed by finance, and every group kind of has their own viewpoint of what's important to them.

And so, as you're you're making these uh, I guess the cost uh increase avoidance of maybe those red flags, how do you know or how do they know what to look for? Is it based on the the years of proprietary data that you already have or a mix of the you know the two with the customer data and your data? How do they how do they know or how do they find out what they don't already know about? Yeah, no, I I I I would say we made a huge investment starting in 22, building out our uh simulation and tracking machine learning modules that are the heart of the platform and when we originally rebranded to sifted AI, uh, that we've recently just made it exposed to our clients through our natural language interface, the uh co-pilot release.

But all that institutional knowledge has been built into our platform so that we can proactively communicate to a shipper, hey, here's what's going on. But some things are blake, you know, pretty straightforward address corrections. You know, if you continue the trend you're on right now, you're gonna have $150,000 in address correction fees. Uh let's get that fixed before you get to that point.

And that's the difference between, you know, having an employee that's got to log on and check versus having something running continuously in the background that's just proactively communicating when there's an issue. And then if there's no news, that's good news. Oh, that's super interesting. So it's not just, you know, oh, I gotta log into another site, you know, I have to create another account, I have to download another app.

This kind of only bugs you when there's something that they need to be paying attention to. And bugs you the way that you want to um be bugged. And, you know, if you just look at the SaaS industry, uh I'm a firm believer that pretty soon most SaaS products are gonna be headless, and you will interface with them through whatever your enterprise chatbot is of choice, Claude, uh Chat GPT. You're gonna come in and you're gonna work out a Cloud and like, okay, hey, sifted, uh um what's going on?

Uh what are my how do my K? I got to go talk to my boss later. What do my KPIs look like? Um you're gonna interface with Salesforce, you're gonna interface with all those SaaS products through something like uh Cloud, and you're gonna expect to be able to get proactive communication as well as interact with it in a very simple way.

And so what are what are maybe some of those light bulb moments for for folks who recognize that they want to get smarter about their parcel spend and they're looking for different solutions. How do they know what to look for to find a company like yours, to find sift it? That is uh that's a great question. You know, you have to have an appreciation and understanding where parcel uh shipping fits in your company profile.

Uh um increasingly it is it's like, especially in e-commerce, like the third or fourth uh largest spend items in a PL. Uh and then uh um obviously we would like you just to come to sifted, but uh um understanding what your options are around parcel spend management, but more as much in the cost savings, having a view to cost prevention. I'm just gonna I'm gonna keep reiterating that because uh um the name of the game is just not to spend the money to begin with. And so uh, you know, as your your or a customer is using the platform, so they they become aware, they they start the onboarding process.

What does that process look like? Uh are you integrating into their existing systems? Is it a new system that they have to set up, maybe some new work processes? Uh walk me through that onboarding process, which I'm sure is different.

It it is. It's pretty simple though. It's it's uh a matter of syncing your carrier data to the platform. Uh we have a lot of different options uh to do that.

We try to make it as easy as possible uh um for our clients. We support, I'm gonna say, 30 different carrier formats. We're actually rolling out uh a new solution that just is universal. It's leveraging AI to make it just uh super easy.

You know, onboarding takes uh a matter of a few days to a week uh then to ingest and get your data and then uh start presenting you results. And then our client success team walks through hey, here's our here are our learnings on the things that you need to pay attention to, we think you need to pay attention to. The platform, when you immediately log in, uh um it provides you recommendations based on our, again, our algorithms, but then we have the ability to customize it to what uh is important to that particular shipper.

Now there's for a lot of you know, sort of the the main use cases for for LLMs is, you know, oh they hallucinate or oh they're not always accurate. You know, there's obviously at the enterprise level, there there's uh, you know, there's governance around your data and and your your documentation. So how do you help those customers who have probably never done any of this before, or maybe they've done it in a smaller aspect, but how do you help walk them through of what makes for good data, how to know when to spot, when when things are hallucinating, and even from the the product lens of how you adjust for the the, I guess the maybe some limitations of like an older Chat GPT model versus a newer model?

Yeah, that's a great, that's a great question. And you know, when I said we tried to do this in 2023, uh those are the things that we just could not uh solve for. And our solution is very unique. I'm not gonna get into a tremendous amount of detail.

That the the LLM that we use in our solution is self-contained within our environment. Uh so when we uh ingest uh a client's a shipper's data, it never leaves our environment. So it is not getting used out in the wild to train models or to do anything. And so that's a really data governance, a really, really important thing.

Our shippers have agreements with their carriers. We have agreements with our shippers, we have agreements with carriers uh around data governance and making sure that it doesn't get used for anything that it's not uh approved for. So that's a that's a huge thing that's built into our platform. The other thing I'd say is that is uh very unique about us, we built a third-party verification process so that when we get uh queries that come through co-pilot, our co-pilot solution independently, we're validating the responses that go back so that we can we're we're confident that the solution that we're giving back to our shippers is accurate.

Uh um and that's maintained by our team. The results of that are maintained by our team so that we can look at the accuracy and make sure that we're we're responding the way that uh we would respond if somebody was using the platform. And now I I imagine for your your target customers, they're almost excited maybe to use some of this because in some of my other conversations, you know, there there's there's pushback, there's uh a hesitancy to adopt these kinds of solutions, or even just AI in general.

But I imagine on on your side of things, it's almost an encouraged and that you you know the the shippers need this help. Is that an accurate assumption? The adoption has been phenomenal. The adoption rate has uh um just been absolutely phenomenal.

I think we're up to 60% now after going live over, I think we went live six weeks ago with uh rolling out to our our entire client base. So the adoption rate has been absolutely phenomenal. And the utilization around um, and there are three components to our uh natural language interface, just asking questions, uh setting up tasks and triggers and alerts, and then sophisticated simulation. And the simulation component, we're continuing to add more capability to.

As you might imagine, there's a lot more uh rigor that goes into that. So you can do really interesting analysis with your shipping data, your contracts, and your own third-party uh data as well. But the adoption rate has just been absolutely phenomenal. And so then that way, uh from an adoption standpoint, is it as simple as, you know, I'm I'm just logging in and I can see all of these different alerts?

Do I get an email? Uh what does that alert system look like? Yeah, you're you're you can log in and see it uh on the um home page, or you can set it up to get push notifications to the platform. You can set it up to get uh email notifications.

And so part of the whole intent, the whole intent is to make the shipper's life easier. Hey, let's set this up for you, and then we'll communicate back to you on an as-needed basis the way that you want to get communicated to. And now for a lot of those shippers, you you mentioned the the 60% adoption rate. I'm sure during this process, you're you're you're getting feedback on maybe improvements or additions that you can make into the platform itself.

Is there maybe anything that you weren't thinking about before, but are or have put it in the product roadmap? It's interesting because AI, this the whole AI push in the industry is educating everybody on what to expect. Some of the things that we didn't expect is how robust and the amount of data that uh shippers were going to want to crunch and that we needed to be able to not we were handling it, processing it was fine, but give it to them in them in an output that they wanted to, and that's extremely variable.

Now, thankfully, the tools in the marketplace are making it easier and easier uh to handle large amounts of data and to graphically represent those. But uh I would say just some things like that kind of caught us a little bit off guard. Well, what what when you find out you're kind of like, well, duh, we should have thought about that. And maybe I if I'm reading between the lines, is a little bit of uh the phrase around token economics or tokenomics, uh how you can manage what you're querying to the co-pilot or to AI based on the type of response or the type of data you're dealing with.

Am I uh Exactly Perfect? Okay, yeah, yeah, because I think I just heard the other day like stop uploading PDFs, you know, to your AI or to Claude, because I kept running into an issue where I was maxing out on my tokens each day. And so I imagine some of that has to be taught to shippers. Yeah, we we uh again, our uh intent is to abstract our shippers from that.

So we manage that on the back end, but it's something that we have to proactively manage again. We we want it to be low friction for and as simple as possible for our shippers. So you it is very rare that a shipper is gonna get a note that you've reached your limit. We've all probably gotten that note.

Hey, you've reached your limit. I love it. Hey, you've reached your limit. You have to wait 17 minutes before you can uh really 17 minutes.

Uh um, we do not do that. Oh, so I would imagine that that's uh almost like a hidden perk for a lot of these shippers that you don't have to worry about, you know, whether you gotta upload a PDF into you know Google Drive and save it as a markdown file or extract the text, or you don't have to do any of that. You you can you know just use your business as efficiently as possible, and we're we will worry about that for you. In the background, exactly.

Perfect. Okay, well Now you can see why it took us a while. It took us a while to get the whole platform architected. Uh so we can handle the data security, we can handle the validation, and we can handle the monetization, the way that that uh AI monetizes the utilization.

And so what are shippers, I guess, able to do after they have that that insight? Is it per cost prevention that you mentioned a couple times, but is there any other maybe hidden aspects that that they might not be aware of? Well, I mean, the the key is uh educating, you know, our our platform, the intent is to educate and alert on what's driving the spend in this space, and then to give recommendations on what they can do uh to manage that, and or give recommendations if uh client experience, their client experience is important.

Uh um, you know, during COVID, you know, cost kind of to a certain extent became less important and prompt delivery became important. So it's it's really about what's driving uh um your need. Is it is it cost, is it client experience, is it gross margin? And then we can uh tailor the platform to help you focus on that.

That makes sense. So customize it per their KPIs of what they're looking for and what that you know supply chain manager is looking to present to their team. Yeah, and it's it and literally could be different for somebody in their comp in the in supply chain than in somebody in finance. Now, is there a particular commodity that this is best suited for?

Is it really just all encompassing of parcel spend? All parcel. And so out outside of that, I think that is about uh well, first before before I let you go, uh is there anything important that you feel we should mention that we haven't already talked about? No, I mean, my my the key things for me is like really uh evolving to prevention from from looking backwards, but preventing things from happening in the future, evolving to, hey, you should we should all, as consumers of technology, expect that I'm buying something that works for me versus me needing to do the work.

And then I've got to feel safe. I gotta feel like my data is safe, and I gotta feel confident in the answers. And I think you have to have a solution like ours where you've invested the time and energy to keep everything self-contained, uh, but still give all the benefits of uh AI capabilities. And you know, we've built a rigorous uh um process that we are doing our own validation as opposed to waiting for uh a shipper to say, hey, I got this answer.

It doesn't make sense. We all see the disclaimer at the bottom of Claude. Hey, it's AI, it makes mistakes. But you know, our intent is for never for a shipper to tell us, hey, I think you got this wrong.

And so it definitely sounds like this is a uh more based on the AI conversations that I've had, that this is your your solution feels more proactive. And then it feels like you're you're taking you know years of you know industry, uh proprietary, tribal knowledge, and then compounding that with the customer data uh to fit their unique needs, which is kind of you know what AI was promised, you know, a few years ago when it was first launched. But it feels like we're finally starting to get you know a lot of that traction.

Is that I assume that's what you're seeing too. Yeah, absolutely. Absolutely. Any uh last words, uh, where can folks follow you, get connected with you, maybe sign up for our demo for with Sifted?

Uh uh come to Sifted.com and uh sign up for a demo, follow me on LinkedIn, follow Sifted on LinkedIn. Uh, we'd love to hear from you. Perfect.

Well, make sure to make it easy for folks, and I'll put all of that in the show notes. But but this was a great conversation, so thank you so much, Sean. All right, thank you. Thanks for tuning in to another episode of Everything Is Logistics where we talk all things supply chain for the thinkers in freight.

If you like this episode, there's plenty more where that came from. Be sure to follow or subscribe on your favorite podcast app so you never miss a conversation. The show is also available in video format over on YouTube just by searching Everything is Logistics. And if you're working in freight logistics or supply chain marketing, check out my company Digital Dispatch.

We help you build smarter websites and marketing systems that actually drive results, not just vanity metrics. Additionally, if you're trying to find the right freight tech tools or partners without getting buried in buzzwords, head on over to CargoRex.io where we're building the largest database of logistics services and solutions. All the links you need are in the show notes.

I'll catch you in the next episode and go check.