Engineering vs Procurement Decision Conflicts

Your supply chain determines whether you can actually deliver what your engineers designed. Most companies don't see it that way. They treat suppliers as a cost center. Optimize for the lowest price. Squeeze for 3% annual savings. Then wonder why they can't move fast when it matters. I see companies with better technology lose because their supply chain can't execute. A defense contractor designed a bracket in G-11 composite to replace aluminum. Better thermal stability, 30% lighter, lower cost. Engineering validated it in 6 weeks. They sent it to their machine shop that specializes in metals. The project sat for 12 weeks because the shop had no experience with composites. Wrong tooling. Wrong feeds and speeds. Parts came back with delamination and chatter marks. Their competitor sent the same type of part to a supplier who works with both metals and composites. They shipped production parts in 4 weeks. The difference wasn't the design. It was a supplier capability. Your supply chain either enables or blocks: ➤ Whether you can execute design improvements or get stuck with suppliers learning on your parts. ➤ If your supplier base matches where engineering is going, not where it's been. ➤ How fast you respond when customers demand better performance. Most companies build supply chains for yesterday's products. Then can't move when engineering evolves or markets shift. You can have the best R&D and still lose if your supply chain can't execute. Execution is the only valid opinion. #SupplyChain #Manufacturing #DefenseManufacturing #Composites #ProductDevelopment

World-class #Procurement Teams aren't debating Build vs. Buy 🙅♂️ ❌ It's Buy vs Buy ✅ I had a lively disagreement recently w/ Oscar Fernández :) which perfectly sums up where I believe sourcing #procurement #tech is heading: One side of the argument: 👉 “We can build this ourselves using AI, low-code tools, and platforms we already pay for.” The other: 👉 “Buying SaaS still makes sense — speed, stability, scale, accountability.” Truth is… both are right. That’s why Build vs Buy is the wrong framing. Today, you’re almost always evaluating: • Buying traditional SaaS with proven workflows, security, support, and uptime OR • Buying AI, infrastructure, and internal capability — and assembling something fit-for-purpose. Neither is “better” by default. They solve different problems. SaaS shines when: – You need speed – You want predictability – You don’t want internal dependency on a few key people Internal assembly wins when: – The use case is narrow – You already have the tools and skills – Differentiation matters more than polish The real procurement tech sourcing mistake? Treating these as philosophical choices instead of commercial ones. Because the leverage isn’t created at negotiation. It’s created earlier — when you ask: • What are we really paying for here? • IP or convenience? • Flexibility or certainty? Buying Procurement Tech isn’t about picking a side of "Build" or "Buy". It’s about knowing when each option makes sense. World-class procurement teams aren't debating “build” or “buy”… They're discussing Buy vs Buy. Thoughts? Drop a Comment 👇

The cost of a part is decided 𝘣𝘦𝘧𝘰𝘳𝘦 𝘱𝘳𝘰𝘤𝘶𝘳𝘦𝘮𝘦𝘯𝘵 𝘦𝘷𝘦𝘳 𝘴𝘦𝘦𝘴 𝘪𝘵. That sounds provocative, but it’s well established in product development research. Decades of design-to-cost and cost-engineering studies show that the majority of a part’s cost is committed during design, not during sourcing. Material choice, geometry, tolerances, manufacturing process — those decisions lock in cost long before a buyer ever negotiates price. This is especially true in automotive and aerospace NPI. These industries didn’t adopt design-to-cost, design-for-manufacturing and assembly, and early cost modeling by accident. They learned the hard way that once a design is frozen, procurement’s leverage drops fast. At that point, sourcing is optimizing around decisions already made. Which creates a persistent mismatch: Procurement is measured on cost, but cost is driven upstream in engineering. So when teams try to “fix” direct materials cost late, they’re fighting physics, not performance. The implication is simple: 𝗜𝗳 𝗽𝗿𝗼𝗰𝘂𝗿𝗲𝗺𝗲𝗻𝘁 𝗼𝗻𝗹𝘆 𝗲𝗻𝗴𝗮𝗴𝗲𝘀 𝗮𝗳𝘁𝗲𝗿 𝗱𝗲𝘀𝗶𝗴𝗻 𝗵𝗮𝗻𝗱𝗼𝗳𝗳, 𝗔𝗜, 𝗮𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀, 𝗮𝗻𝗱 𝗻𝗲𝗴𝗼𝘁𝗶𝗮𝘁𝗶𝗼𝗻𝘀 𝘄𝗶𝗹𝗹 𝗮𝗹𝘄𝗮𝘆𝘀 𝗱𝗲𝗹𝗶𝘃𝗲𝗿 𝗱𝗶𝗺𝗶𝗻𝗶𝘀𝗵𝗶𝗻𝗴 𝗿𝗲𝘁𝘂𝗿𝗻𝘀. That's why LightSource starts with the BOM and engineering context because that’s where cost, risk, and supply decisions actually originate. When sourcing is connected to design intent early, procurement stops reacting and starts shaping outcomes. 𝗖𝗼𝘀𝘁 𝗶𝘀𝗻’𝘁 𝘀𝗼𝗺𝗲𝘁𝗵𝗶𝗻𝗴 𝘆𝗼𝘂 𝗻𝗲𝗴𝗼𝘁𝗶𝗮𝘁𝗲 𝗮𝘁 𝘁𝗵𝗲 𝗲𝗻𝗱. 𝗜𝘁’𝘀 𝘀𝗼𝗺𝗲𝘁𝗵𝗶𝗻𝗴 𝘆𝗼𝘂 𝗱𝗲𝘀𝗶𝗴𝗻 𝗳𝗼𝗿 𝗮𝘁 𝘁𝗵𝗲 𝗯𝗲𝗴𝗶𝗻𝗻𝗶𝗻𝗴. How is early procurement is involved in your NPI cycles today?

𝗬𝗼𝘂𝗿 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝘀 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗯𝘂𝗶𝗹𝗱 𝗶𝘁, 𝘆𝗼𝘂𝗿 𝗖𝗙𝗢 𝘄𝗮𝗻𝘁𝘀 𝘁𝗼 𝗯𝘂𝘆 𝗶𝘁 - 𝗵𝗲𝗿𝗲'𝘀 𝘁𝗵𝗲 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 𝘁𝗵𝗮𝘁 𝘀𝗲𝘁𝘁𝗹𝗲𝘀 𝗶𝘁 𝗶𝗻 𝟯𝟬 𝗺𝗶𝗻𝘂𝘁𝗲𝘀. Start with one sentence: "We need [capability] to [business outcome]." If you can't write that sentence, you're not ready to decide. Then rate on five criteria. Time to value. Total cost (capex plus opex over 3 years). Strategic fit. Risk - lock-in, security, data residency. Internal capability to maintain it. Now the uncomfortable question: "Can we really maintain this?" Most teams overestimate their ability to support custom builds. I've watched a team of four build a beautiful data quality layer. Six months later, two engineers left. Nobody could touch the code. Force a choice by asking: "If we pick Build, what must be true? If Buy, what guarantees do we need? If Partner, who owns outcomes?" The answer changes as your team matures. What you should buy at 20 people, you might build at 100. Review this every 6 months. 𝗪𝗵𝗮𝘁 𝗱𝗮𝘁𝗮 𝗰𝗮𝗽𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗮𝗿𝗲 𝘆𝗼𝘂 𝗰𝘂𝗿𝗿𝗲𝗻𝘁𝗹𝘆 𝗱𝗲𝗯𝗮𝘁𝗶𝗻𝗴 𝗯𝘂𝗶𝗹𝗱 𝘃𝘀 𝗯𝘂𝘆 𝗼𝗻?

Procurement focus on OpEx 5x more than CapEx. Or get involved so late in CapEx it’s like lipsticking a pig. This can be dangerous to a company’s bottom line. So, how do we fix that? Let’s look at the differences first: CapEx Procurement: ➟ Purchases are often one-offs ➟ Large Purchases & for long-term use ➟ Likely custom made purchases OpEx Procurement: ➟ Weekly / Monthly payments ➟ Everything from software, raw materials to packaging ➟ Repeatable purchases How should we approach CapEx more effectively? 1. Don’t apply processes based on repeatability to CapEx CapEx is a one-time purchase and often something unique. Procurement teams working on CapEx projects need to focus on specifics: e.g. that all requirements are met, the budget & delivery timeline are delivered. 2. Track costs associated with change orders long after the contract has been agreed. Change orders can be a big part of Capital Expenditure and often even the scope changes long after the deal has been agreed. Procurement should ally with finance to keep track of the longer term financial outcomes. 3. Don’t muddy the waters between recurring costs and the one-time investment Specifications for CapEx are rarely “off the shelf”. You’re sourcing a unique product or deliverable. But some CapEx projects may have OpEx applicable components. E.g. a new building may well have recurring expenditures (e.g. network, maintenance, security). Keep visibility of these by separating them out instead of having them rolled into a blanket invoice. --- And here’s how Procurement can really add value in the sourcing process: ✅ In one major CapEx investment I supported 6 years ago, we put the emphasis on multimedia (blueprints, drawings, schematics) and full days of workshops and tours with all the bidders together in one place. This better communicates the requirements, helps the bidders ask the right questions and streamlines your time (not having to contact each supplier individually) ✅ Leave room in your RFP for true flexibility. Suppliers who deal with CapEx projects are often highly specialised and may well be in a better position to advise on the best approach to take than anyone from within your organisation. ✅ Make sure you have a detailed breakdown of costs in the bid responses. This helps proper analysis and supplier comparison. It also enables you to focus on the most expensive components to reduce the overall project expense. ✅ Use predictive insights and client reference calls By asking the right questions you can get a good barometer of how often and how much the change requests might cost. ——- What do you think of this advice? What would you add? Let me know in the comments. Hope that helps! Feel free to tag someone in your team who could find this helpful or ♻️ repost it to your network.