Transcript

Speaker 1  0:05  
Of course, thank you for joining me right after your lunch. So it's best to have a presentation after lunch. I think everybody still has their energy, so maybe just to start by a show of hands. Who in this room is not a wire harness manufacturer? It's not okay. So just throw up. What are you guys doing here? Let me do the automotive and your OEM consumer, OEM fire trucks, OEM

Speaker 2  0:52  
forklifts, OEM, anybody else,

Speaker 1  0:58  
machinery OEM, anyone else? Okay, I love that we have OEMs in this room, because quoting is not only a problem or challenge for wire harness manufacturers, and I think that the more of the ecosystem that is familiar with this, the better off the industry will be. So, we're giving a talk about breaking the manual quoting bottleneck, and sometimes quoting feels like breaking your neck, so hopefully we can, we can make some progress here. And just to start, I am not a software vendor that's just here to pitch you a product. I'm a guy that owned and ran a contract harness manufacturing shop for 35 years. I sold my company to Aptiv in 2021 and then I built a tool that I wish I had when I was manufactured, and what I'm going to share with you today is the playbook that I wish I had when I own my own business, so as I said, I've spent 35 years in the industry. I owned a company called Alcom Systems. We were a wire harness manufacturer, primarily for aerospace. I am also a board member of WAMA and co-chair of the WAMA IAT, the innovation advisory team under WEMA, and I'm the founder of Cable Tech and CEO of Cable Tech, and so the data that I will be talking about today, it comes from three places. First is my old shop, my own experience, and what I saw in my, in my time at Delcom. Second is the IAP, the organization I the web organization, that is sort of advancing innovation. We did a survey last year, 42 manufacturers, and we're pulling out this data. Some of you have seen it already. And then live customer data, talk to customers every day, so the data I'm sharing with you here, it's not a theory, it's real. And so, when I cite a number, it will be from one of those resources. It's not going to be perfect, but it's the best data that we've been able to get so last year, as I mentioned, Rama conducted an industry survey, and the numbers - they're not subtle. Three out of four shots describe their own quoting as being manual, time-intensive, and too slow, and this is their words. It's not mine. More than half of them admit that they depend on tribal knowledge specifically in their court process, and then the scary part over around two thirds of the people in this room are quoting in Excel. Of the people quoting in this room, is anybody not quoting in Excel? Can you throw out what you use for quoting? It custom built

Speaker 3  5:02  
our ERP system has

Speaker 1  5:05  
ERP system. Yes, custom built. Okay, so three people in here are using custom built or an ERP, and the rest are coding in Excel, a great majority find design errors after quoting, 83% find design errors after quoting. We benchmark quoting time and quoting around time, so the average low touch time when a fairly simple quote is five hours, meaning five hours of your engineer and supply chain time. That's for a simple build. There were some automotive OEMs here, or off-road systems. Know that it could be 20, 3040, hours to coat a single harness, and then the average turnaround time for a quote is two to three weeks, two to three weeks, just to respond to an RFQ in 2026 That's that's not right, and that's certainly what we should be solving. So quoting really isn't just a workflow problem, it's also a PNL problem. It affects everyone's bottom line in revenue when you take two weeks to quote, but the customer needs it in two days, and somebody gets the quote in two days, the order goes to whoever quoted it first, not whoever quoted it best. So you could have spent two weeks on something, and had a great result. However, it took too long, and you lost the opportunity, right? That's not money, that that's not an investment, that's a waste, you margin side, you've got tribal knowledge in my former company. I often, on large bids, I often gave the same quote package to three or four different engineers, because I was frankly nervous about bidding incorrectly, and guess what, I would get three or four different quotes, and then I'd have to figure out whom do I trust, because it's really not a democratic process, right? And then we've got a talent problem, we've got our best people, our best engineers, those tribal knowledge guys. They know everything about the plants, they know everything about the customer, right? They know who needs what and what's not stated on their drawings. They know what are the right parts for a certain customer. What do they do? They're spending their time chasing missing parts, looking for specs, finding an alternate that's a hiring problem disguised as a process problem, and the kicker is that 76% Again, this is not my data. 76% of manufacturers find designers after they've won the job. So, you bid it one way, and then you discover that you bid it wrong. Okay, so the bottleneck there doesn't, doesn't just slow down the sales process, it guarantees a production loss, because if the product stops in production, it costs you money, and if you've got to buy new parts and scrap others, it costs you money. Sure, maybe you can transfer some of that cost to your customer, but that's not a fun negotiation. Right? Earlier this year, I attended the WAMA annual conference. Was anybody here at the WAMA? Not here in Las Vegas at the annual conference. Only one person. Yes, Jen. Cameron, any more? Only a couple people. It's a great conference to attend, by the way. And so we asked the question, and I'm going to ask you now, How many of you have your most expensive engineers? Is doing data entry, typing bombs into Excel just to get a quote out. Can you raise your hand if you've got your top engineers just data entering a lot around quotes? Okay, over half the people in the room, and we call this being data archeologists, right? They're digging through the artifacts of the PDF, they're digging through the specs, they're going online looking for the information on the part only to reconstruct what the customer wants that he didn't tell you in the beginning or the data that you don't have, and so this is that really a sustainable business model in 2026 when we've got so many tools and we've got to operate so, so quickly, so efficiently, and our customers depend on us, is that the right business model? We know that OEM send incomplete PDFs to you, they're missing information, terminal seals, plugs, they're missing for those OEMs in the room, you're missing terminal seals and plugs in your design. You don't tell us what are the right wire lengths, so the coverings, okay? It's really hard work, and sometimes we don't get it right. And engineers, they have to spend a lot of time looking at that data, just on the quote, they should be spending the time on the manufacturing floor, figuring out how to build it faster, how to build it better, how to automate, and then, of course, when the manufacturers find a problem, they can't make a decision on their own, right? They got to flag a problem, they got to go to the OEM engineer, they got to get approval, that all takes a lot of time and costs for both entities, so the IT right, the Innovation Advisory Committee identified four root causes for this, and these are compounding, compounding causes. The first one is incomplete customer drawings. 50-7% say this is their single biggest blocker. My former business at LCOM, I probably had around $10 million on hold at any given time, $10 million on hold, waiting for customer design instructions, and those were typically issues related to the customer design. Right, that's a lot of money, lot of working capital, a lot of unplanned production that we had to absorb, of course. What that means is we're missing parts, we're missing seals. We didn't have the right material in stock. We will email the customer. Three days are passing. By the time we get the information, this whole thing starts drifting, and it's not really the OEMs fault, it's not your fault, guys. The tools that you use don't make it easy for you, and we'll come back to that in strategy form. Then the second problem is the way data is exchanged between the OEMs and their suppliers. 90% of the data exchanged is in PDFs. Okay, 2026 90% of the technical data being exchanged between OEMs and manufacturers on something as complex as wire harnesses is in PDFs. What are you guys supposed to do with that? Right, the data, the original data exists, usually. Okay, it was created in CAD. Yes, I know some people created on a napkin, some created on Visio, but typically OEMs, larger ones, more sophisticated ones, do have CAD tools available to them, and it's digital data and has a lot of value, but you choose not to give it that way to your suppliers. You choose to print screen, run it to PDF, and send that data, so that they can recreate this data when they get it. And so, what happens with the manufacturers? They get the PDF. It might have a bomb, it might have information all over that drawing, and they type, re-type every line by hand. Most don't use any kind of tools. The OCR sometimes work, sometimes doesn't work, especially on large drawings there. Not really effective, and so they are retyping all this data. Not only is that inefficient, it often leads to mistakes. So, really, for everyone, both OEMs and manufacturers, if we can fix one thing this year, the ideal thing to fix is how we get the right data to the manufacturers. Let's solve the PDF problem, and we'll cover that in strategy two. The problem number three, root cause number three, is really the challenge of data, and I'm talking right now mostly about commercial data, but it's also, of course, technical data. In a typical quote, when we look at a 50 line item bill of material, that's not huge, it's a medium-sized build. In a typical quote, we can have over 1000 options of where to buy the materials or what are the right materials to buy. Wires aren't specified by a manufacturing part number, they're specified by a spec. I can buy a different wire from IWC and a different wire from AWC, and it should meet the spec, right, but there's all these different combinations. Let's not talk about colors, and then same thing with terminals, and even if it's the same part, there's so many different suppliers, and then we've got to deal with quantity options, so you get an RFQ, and they want you to quote 525 50 102 50, right. Oh, it should be really easy, but it could be that a five I'm buying from one supplier and a 500 I want to buy from another supplier. That's a lot of data to one get. How do you get all this data? Of course, most of it's done pretty manually, and then even if you've got an amazing sourcing team and they get you data from all of the suppliers, they've got to put that back into their Excel or their ERP or the custom quoting system, and that's got to figure out what's the right combination, it could be price, it could be delivery, it could be the people we like to do business with versus others, so really, How is it possible? How can you expect their quoting person to make the optimal decision within a reasonable time frame, and you know what? What typically happens, right? We've got.. we all.. I had a coding team, you, many of you have coding teams, and they're really experienced people, and they've done it for a long time, and we trust them, but they're humans, and they can't make a decision across some, so much data, so often the first quote, the first person that replies, just like when you reply to your customers, the first supplier that replies is who gets on the costing sheet, and that might need that, that might not be the best supplier, somebody else may have quoted three or four days later, that doesn't make it to your costing sheet, that might be a reason you use the business, because you didn't know that that got materials available in two weeks, or you use the $3 component when your competitor knows to buy it for $1.50 we see this happening every day, you and then root cause number number four, and this should be one that keeps all of you up at night, is that tribal knowledge again. 57% of the responders said, quoting depends entirely on a small group of their engineers and sourcing people, and to onboard a new engineer takes six to 12 months, and that's a junior person, like they don't know everything about your business in six to 12 months, but just to get an average person in the door that stays in the door and that can do some meaningful work takes six to 12 months. So, what happens when your best estimator goes on vacation for two weeks? Are you still able to quote, and typically the answer is yeah, of course, you know we are going to manage because we have to quote, but is it the same level of quality, is it the same level. Of understanding, or are you some of you, as the business owner, staying over the weekend to do those quotes, because nobody else is available to do it, so that's a single point of failure on your entire revenue stream. The quoting is the gateway to the customers, right? It's what drives your new revenue, and it's hinges on a very few amount of people in the business, and of course, those estimators that are experienced, some of you are in this room at some point they retire or go to a finer job, right, and it's really hard to start up again. It feels like, you know, building the business all over, it's quite painful. So this is a huge point of failure that that should be alarming to everyone, and of course, those four causes that I mentioned, they compound. We see that two hours, we believe two hours is what a quote should take, that same five hour quote, it can be done in two hours, and I've got this for you, it can be done in less than one hour, and I'm talking about that five hour quote with 50 circuits, not the 300 circuit quote, that certainly takes more effort, but that's the industry average. Two to three weeks is what the industry average is from RFQ to delivery of a quote. Is anybody here turning quotes around consistently in less than five days? Consistently, one, two, are you doing it in less than three? Keep your two people here under five days. Are you doing it in less than three days? Yes. Are you doing it in one day?

Speaker 4  22:24  
Sometimes, sometimes,

Speaker 1  22:25  
sometimes. How many quotes do you process in a month? 1520, How many do you process? 1520, So, if you're quoting 1520, you can do it quickly. How many people here quote more than 100 quotes a month? Right there, no, none of you quote more than 100 100 quotes, 100 assemblies, unique assemblies a month. Okay, we've got some more hands, right? We've got people that quote, we see people that quote even 1000 a month. It's not my former company. We quoted more than 1000 unique assemblies a month, so that volume process and make those decisions certainly compounds. So, you know, what do we do about it? It we're going to talk about four proven strategies, and none of these require buying new people. Okay, so we're not asking you to hire new people to solve this problem, and we can, we can improve on, on, on the challenge. So, first, pull the data out of the PDF. Stop retyping. In fact, it's almost like take away the keyboard, because if they've got the keyboard, they're going to type, and so we are fortunate to live in a time where OCR has improved, and, and more recently, what we call AI-driven extraction extraction, where we can read the customer drawings the same way that an experienced estimator would, and this is really important. It's not just reading the drawing, it's reading it in the same way that an experienced human estimator would, which means, of course, looking at the bound table, but also looking at the connector tables that are all over the drawings, looking at the wire lists, the front two lists, identifying the revisions, looking at the notes, and also identifying. What we don't know, we know the wire color, and we know the wire gage, but they didn't tell us if it's TXL or GXL. That's a question, happens quite often. So, identifying what we don't know, using the tools to work in the same way that the human would, ideally being able to flag things that are missing, like those missing terminals, missing seals, or accessories, even before the engineer has to open the file and identifying certain workmanship standards that one customer has over another that tell you that you need a certain tape when we're using this particular connector or lacing ties over cable ties, like all this information that it lives in the heads of the few people that can be codified into a tool that interprets the drawings similarly to what the human engineer would, I want to caution here, because it's not going to be perfect, and we certainly know we spend a lot of time benchmarking data, and we certainly know that it's not perfect, and so we shouldn't. I'm not saying here that the AI is going to replace the people doing the work. What I'm saying is we use the AI to be that data archeologists, right? We use the AI to do the heavy lifting, and then the engineer needs to stay in control. The important thing is that the engineer now has time to ensure that they're focusing on the right things, so that the errors that we catch aren't found on the production floor, they're identified on the front end, the result well before, and dealing well before we quote. So next is the tribal knowledge, right? I already mentioned there's a lot of tribal knowledge that goes into drawings. First, is this particular customer wants things done one way, another customer wants it done another way, right? These certain things are missing from the customer drawing, and this particular customer specifies them another way. One of my customers would identify the coverings, the shrink sleeve and expando, and they would just call it XX. You guys seen the XX coverings before, yeah. What does it mean? Whatever fits right. So, think about a situation in certain industries. It's not a big deal. In aerospace, it's really hard. We were using Fire Sleeve. Fire sleeve had a 12 week lead time and cost something like $10 a foot, and then you always have to have it the smallest it fits, it's not really whatever fits, it's a smaller size it fits, so invariably you know we're we don't have the right one in stock, we have one harness has 12 different sizes, that's the tribal knowledge, and then ideally we can get to DFM tools, validation tools at the coding stage. The back shell does not fit the connector, so okay, that happened quite a lot in my former business. And finally, there are material rules, there's logic to the materials being selected, but the OEM doesn't always apply that logic, and you know there's, there's a few OEMs here. It's not the same guy designing all of the harnesses for the specific platform, or what you're quoting. It could be three or four guys that were using two or three different CAD tools or libraries to quote to design, and that's where you got to quote, so we can apply certain logic to the materials that helps to optimize that and that makes a real real impact I. So strategy three, whatever you invest in the quoting side helps you prevent fighting it in production, and this is a really challenging balance, because when you quote, you're always under rush to quote, right? They want it as quickly as possible. We already said it takes two to three weeks. Some of your competitors can do it faster. There's an incumbent supplier, and he has the quote in already, and you try, you know, your sales person to be working for six months to get an opportunity from this customer. Finally, they say, "Here you go. That incumbent supplier quotes it in a day, and, and the buyers like, hey, your company is no good thing, you promise me you're going to do a great job, but I've been waiting 10 days, I didn't get the quote right, so you're under huge pressure to quote quickly, you don't win everything you quote, most people don't win even half of what they quote. The industry average is 20% so 20% of what you quote, you win, which means 80% you don't win. So, so we're losing money, the more time we spend, the more money we're losing, but on the other hand, if we don't find certain things in the quote and we win that 20% or maybe we're better, we're winning more, we're probably not going to find it in the contract review when we get the PO, we're going to assume it was taken care of, and then the floor finds it, because the floor should always find it. Sometimes, sometimes they don't, it makes it into the end product, and that's a real problem, but assuming in many cases the floor finds it, we've got changes, we've got retrofits, we got rewards, we got new materials, we got internal work order changes, we gotta get procurement to realign, that's a huge cascading effect. So, what you find on the front end is a value, and in my experience in my business, when we identified things in the front end and our customers learned that they can trust us to find their gaps, they gave us more business, right? We became not just a part supplier, a commoditized part supplier that's competing on price, they wanted to give us the work, and of course the real problem is nobody measures anything we might measure the efficiency on the floor to some extent, but we don't measure that front end of the process, and so before you fix anything, you've got to measure it, and I can't tell you how many conversations we have with companies, and we say, how many assemblies do you quote, and we go, we don't know, and we ask him, what is the turnaround time? They typically say it's about two to three weeks, but no, what is it? Is it two weeks? Is it three weeks? How much time do you spend on quoting? Almost nobody has that data, and this is the front end of your business with your most precious, costly, and knowledgeable people, and it's not even better. So we certainly recommend that you start mapping and measuring these things, and the touch time and turnaround time are two different things. Just like in production, you're measuring both how long it takes you to deliver and how much time people spend on the product. The same thing should be with any key processes, and certainly with quoting, and then if you have those measurements, we're happy to share what benchmark industry benchmark looks like. So again, the IAT, we are capturing this data. We've got multiple members in the group that are contributing to this data, they're very large wire harness manufacturers, to really small, and we're benchmarking this data, and then we, as cable tech, are certainly benchmarking this data with our customers and our system in a more automated way, so we can certainly provide. Anonymized industry feedback to what it looks like. Are you doing better? I mean, a couple of companies that I can do quotes in one day. So, are you doing better or worse than your peers? And how much time is that? Is that taking you? And I just want to emphasize, we can't fix the problem unless we measure it. It's all intuition at that point, and when you know when we start measuring, then we can start seeing what has the impact. So, we've got all kinds of customers. I can't list all of them, but I'll just mention a few that that have told us and have said it publicly. A company, KCM Cable Small Wire Harness Harness manufacturer Corey. He said it took me 18 hours to do a quote like this. He's the guy quoting, he's the owner of the shop, so that we'd spend my weekend on this, and now I went from 18 hours to 45 minutes. He got his weekend back, Resco Electronics. Some of you might know they have a 30 day backlog. 30 day quoting backlog is now down to two to three days. That means they can go after more business because they're not quote backlog, right? They can respond to customers, S and Y, double their output, same same amount of folks, and there's many, many other stories, right? It does take time. These are examples of three manufacturers, usually we see results in three months. Certainly, we see this kind of repeated results in six months, and our targets are 80% in cycle time reduction, meaning in turnaround time reduction and 50% in your direct touch time, and we're confident that this is achievable. So three concrete things you can do when you go back to your shop on Monday, some of you will be back on Friday. So, the first pull your last 20 RFQs, see if you can get any data. You don't have to start forward, but look at, look at when that quote was received by your sales team and when it was delivered, everything else doesn't matter. It went into the sales inbox on the first of April, and it went out to the customer on the 24th That's the measurement, and that's the deal that you, you know, end up losing, because it took too long. I think you will. You will see a lot when you look at that second, find where your most critical and single point of failure may be. Who is that person in the process that is the most critical in that process, and then look at what happens if they're done for a week or longer, or they leave right. Do you have a process to support that? And then third, pick one of the root causes that I shared with you, and I'm happy to share the slides. I think they'll be available afterwards. Pick one of them and fix it this quarter. These are fixable things. If you spend time on it, we certainly can help you if you're interested, or if you want to do it alone, we're happy to advise you on that. Just in terms of general feedback, don't try to fix the entire process. It's hard to go from three weeks to three days, but if you certainly want feedback from members of the IIT, the Wema IIT, or if you want feedback from Cable Tech, we are happy to provide that. So our vision, and this is what a perfect Tuesday should be. In our industry, right, is that the OEM engineers send you the manufacturers rich CAD data. Your smart quote engine processes processes instantaneously. You're immediately able to reply with accurate pricing, both in terms of materials and labor, based on your equipment and your machines and your standards in your plant, and you reply back to your customer in hours, not weeks or days, and that's not science fiction. This is something we're working towards. We've got huge initiatives, both as an industry, Wema, and companies like table tech, there are others that are working to support this problem here at the conference, but the important thing is, if you're not addressing it now, your competitors are addressing it, and you will be at a disadvantage, because that's where the OEMs are moving to, and that's a competitive requirement to respond in the same way and in the same standard. So, you've got to choose, do I want to be a leader or am I going to wait for my customers to compel me to catch up, and before opening it up to questions, I do have a few plugs. First, on your table, we've got cards that are free wire harness tooling search. Okay, it's a, it's an open website. You just need to register your name. You don't need to be a customer of cable tech. You don't need to pay for it. It's available to everyone. It's our contribution back to the industry. We spent years collecting data. It's not perfect, doesn't have every part that you might have, but it's got quite a lot. And feel free to use it. Feel free to tell us if it's missing anything. We were happy to look back. If anybody don't have to just take the pictures, we have cards that you can scan. You bought here in the purple hat behind me has these cards, take it, have your team use it again, and give us feedback. And then we are here at EWTPE. This event is is done by the organization called WEMA Wire Harness Manufacturing Association. How many here are members of WEMA? Around half, which means half are not yet members, and so this is an important organization. It's our industry. Certainly, I wasn't a Whamma member all of my years in the industry, but now that I am, I can say it's a great organization, gives us a lot of benefits, we're learning a lot from other experiences and best practices. There's a lot of important educational materials that the organization shares, connecting us with suppliers and customers, and it's a great network, and even have built friendships around this industry, you know, common hardworking people solving similar challenges. If you're not a member, visit Buena at their booth, I think it's 829 and they do have discounts on their membership. Thank you. Any questions, comments? Yes, the tool search does not have an API. We are not going to expose it via an API, we will have it within our cable tech application to automate the identification of tooling when we see components that require tooling, including finding. Identifying where it is and how much it costs during the time of quote, but for this particular tool it's a web lookup, and we're going to keep it that way. There's a lot of work we spend to build this data, and we wanted to capture that work.

Speaker 4  45:21  
A big part of the you talked about is using data rich design inputs into the quoting process. How do you see the benefits of just doing the quoting process data rich input versus the whole system data rich input? So using the data rich format into all of the manufacturing, engineering, the form board design, the sub-assembly instruction, the tooling instructions, all of that.

Speaker 1  45:48  
Well, certainly, that's the biggest benefit, is to get the data rich all the way through the manufacturing process, assuming that the fidelity of the design would support that, because quite often what we see, even if we can get an XML file, a lot of things were not defined in that in that engineering file. A lot of things are left for the manufacturers to interpret and figure out. I know you do your own designs, and when you do your own designs, and you know it's going into the manufacturing process, you certainly understand the value of doing it right, but not all of the OEM engineers see that same value or have the same skill set, right, the same training that you have when you are using these tools to do it at that level, so the goal is certainly to go from design to quote to the manufacturing floor to your system to your Comex to your series testers, or what you're using, that's going to require a lot, a lot of work. It's not just,

Speaker 5  47:02  
it's

Speaker 1  47:03  
not just the data exchange. Do

Speaker 4  47:05  
you see a benefit to driving a common wire harness design language or wire harness design format?

Speaker 1  47:14  
Good question. So, the question here was, is there a benefit to having a wire harness design format that's accepted by first of all, you guys, system manufacturers, that you can read that you don't have to pay a $50,000 license for in order to read that data or do anything with it, absolutely, and we do have several committees at WAMA that are focusing on this and I think you would be a great member. We have a data exchange committee that was recently formed to try and solve this for the industry in a practical way that has both an education component, a practical understanding of like what's the right checklist of what are the things required to be in the design, and then that design data exchange protocol, and we've looked at KBL and VC, which are automotive, which are really good standards, but they're really large models, may not be practical for industry, but if you're interested to participate in that, we've got active initiatives, and we just had to talk about that earlier today as well, and we love to have people that are passionate and really did understand that it's pretty complex. Any more questions? So, what are you going to do on Monday? Analyze your quotes, benchmark, look at how much time it's really taking you, and figure out if that's that's acceptable to you.

Speaker 6  49:16  
All right, enjoy the rest of your afternoon