Why PPAP Is the Key to Scaling Additive Manufacturing

Solutionology Podcast Episode #32 | 22:05

Description

While highly responsive, additive manufacturing carries more variability than traditional methods like injection molding or machining. Without discipline, that variability can create significant risk.

DI Labs has spent years supporting mission-critical industries like medical device and defense, where precision and reliability cannot be compromised. Our work ranges from accelerating clinical builds to developing full-scale volume production programs. Through it all, one lesson is clear: additive manufacturing does not scale without a disciplined approach to risk. Our Production Part Approval Process (PPAP) is the risk management tool we use to align with clients on what matters most, prove results with iterative samples, refine QC priorities, and clarify the important balance between timeline and budget.

This guide distills a recent Solutionology Stories Podcast episode where hosts Carl and Brian Douglass, Co-founders of DI Labs, offer guidance designed for product developers, managers, and innovators who are building a path from concept to commercialization.

1. The double-edged sword of additive

Additive manufacturing can move fast, which is a strength, but that same speed also creates risk. When teams rush forward, they can miss critical aspects of the part, skip dimensional checks, or underestimate the variability that is always present in additive processes. Those gaps can show up later as scrap, yield loss, parts that do not perform as intended, and lost time.

Speed increases the chance of missing critical dimensional details
Direct-to-build approaches can bypass the alignment step on what matters most
Inherent variability, if unmanaged, creates yield loss and scrap

“Additive manufacturing is a double-edged sword. One edge is speed and innovation. The other edge is risk.” — Carl Douglass

2. What PPAP really does

PPAP has different flavors, but the goal is always the same: alignment with clients and becoming familiar with the products before scaling. By running samples, checking dimensions, and experiencing the part together, we can collaboratively establish the right risk management controls for the process.

Clarifies what success and failure look like for the client
Identifies and agrees on which dimensions are critical
Uses iterative runs to validate results before releasing to scaled production

“We want to become familiar and experts at the products that we are making, before we scale into thousands of parts.” — Brian Douglass

3. Practical risk management in AM

Every program faces different constraints. Some are driven by fixed deadlines like clinical trials, where there is no second chance. Others are driven by cost or by the need to make progress quickly. PPAP provides an adaptable structure allowing for customization, sometimes accelerating steps, sometimes overproducing critical parts, and sometimes taking more time to validate.

Clinical programs often require accelerated PPAP or intentional overproduction
Skipping PPAP usually results in scrap and wasted resources
Programs with more time benefit from iterative sampling and dimensional alignment

“If you do not have time to do it right the first time, how are you going to have time to do it right the second time?” — Carl Douglass

4. The ROI of doing it right

Spending time and money on PPAP up front may feel like a cost, but it prevents much larger losses later. Even a modest validation step can save multiples in avoided scrap, rework, and program delays. The math is straightforward: PPAP returns far more than it costs.

A $5,000 PPAP run can prevent $45,000 in wasted parts and rework
ROI often ranges from 10 to 100 times the investment
Early validation avoids costly resets and keeps programs on track

“An investment of $500 or $5,000 on the front end can return 10 to 100 times that amount on the back end.” — Carl Douglass

5. Collaboration is non-negotiable

PPAP only works when both sides are engaged. Without an active feedback loop, even small variables like geometry orientation, build location, or calibration differences across machines can create major variation. DI Labs drives repeatability by controlling what we can and making those variables visible to clients for maximum alignment.

Alignment on critical dimensions and tolerances is essential
Orientation, build design, and processing parameters need to stay consistent
Calibration differences across machines must be understood and managed

“Without a bidirectional feedback loop, where expectations are set, understood, and measured against, you do not converge on a solution intentionally. You just get whatever comes out.” — Brian Douglass

6. Tailor PPAP to the program

There is no one-size approach to PPAP. The process details are driven by the manufacturing method, number of parts, complexity of the product features, and application requirements. It can range from simple visual checks to extensive dimensional validation and strict repeatability controls.

Ergonomic or low-risk parts may only need surface checks
Dimensional-critical assemblies require full validation and iteration
High-volume programs may require locked orientations and consistent builds designs

The goal is always the same: scale the PPAP process to fit the program, balancing time, cost, and risk in a way that creates confidence.

Final thoughts

PPAP in additive is not about slowing innovation – it’s about scaling it with confidence. The practical move is simple: align early, invest a little up front, and avoid the expensive, unnecessary rework. DI Labs sees its role as guiding teams through that journey, helping them anticipate risks, control variables, and deliver parts with confidence at scale.

DI Labs PAP Process: https://dilabs.cc/production/

DI Labs specializes in additive manufacturing for volume production in mission-critical industries like medical devices, defense, and consumer products. With over a decade of engineering and product development experience, DI Labs delivers repeatable quality, reduces risk, and helps teams balance time, cost, and regulatory requirements as they scale additive into dependable supply chains. Learn more at www.dilab.cc

Chapters

00:00 – Introduction: Risk and discipline in additive manufacturing

01:05 – The double-edged sword: speed vs. risk

02:12 – What PPAP really does in additive

07:35 – Managing trade-offs: deadlines, budgets, and learning “in the heat of battle”

12:04 – Why doing it right the first time matters

12:37 – The ROI of PPAP and avoiding costly rework

16:28 – Why collaboration and feedback loops are non-negotiable

21:02 – Tailoring PPAP to fit the program

22:25 – Final thoughts: guiding teams from concept to commercialization

Speaker:
Welcome to the “Solutionology Podcast.” We’re here today to talk about production using additive manufacturing, and more specifically, understanding and managing risk. A lot of this conversation is centered around a PPAP process. PPAP stands for Production Part Approval Process. It’s all about scaling the process for production launch.

Speaker:
So risk in production means something a little bit different for everyone. And on this podcast today, we’re going to talk about what managing risks and being aligned on risks means as parts are going to some production state. That could be a clinical trial, it could be full-scale series production, it could be batch production where the part changes through each production cycle. But we’ll talk about risks.

Speaker:
It’s like additive manufacturing’s a double-edged sword. One edge is all about the benefits, the speed, and what you can accomplish with additive manufacturing, and the other side is the risk. Because of the first positive aspect of additive, we can move much faster, and we can take a digital file and get it straight into production. Because of the speed available, that in and of itself presents risks of missing key aspects of dimensional control and what’s critical about the part. And then on top of that, we’ve got other risks associated with the variability inherent in additive manufacturing. You could argue that there’s greater variability in additive if you’re not paying attention.

Speaker:
So when we’re thinking about risk mitigation for our production projects, a key component is alignment with our customers, collaborating with them to understand what their expectations are, what success and failure look like for them, and then doing the best we can to incorporate a PPAP process to manage or mitigate those risks and keep them below the client’s threshold.

Speaker:
PPAP stands for Production Part Approval Process. It’s a common term that’s used in many manufacturing verticals. It does have different flavors, so there are different ways to apply a PPAP process, and we’ve got our own. What does that look like for us?

Speaker:
It’s about understanding the application, understanding the requirements, and then, when possible, we want to do iterative steps as we are gearing up for production. We can have samples and dimensional controls put in place. As we both experience the product our client needs, we need to experience it running through our machines and see what the results are. We also need to provide them with samples so they can see results. As we go through that, we agree on what dimensional controls are right for this. And as we add dimensional controls, it becomes more challenging to make, and often parts are dropping out quickly. So we want to agree on that dimensional aspect.

Speaker:
When we say “dropping out quickly,” you’re talking about yield loss and scrap, right?

Speaker:
Yeah, it’s a big consumption of resources as we’re going into production.

Speaker:
Why even do a PPAP process though? Everybody talks about additive as this magical thing that builds parts up from nothing. Why not just produce parts and expect them to come out the way you’ve designed them?

Speaker:
Well, the PPAP for us, we want to become familiar and experts at the products that we’re making. If we have an order come in for 1,000 parts or 1,000 assemblies that we’ve never seen before, and we don’t get a chance to experience it, become familiar with it, and put controls in place so that we can monitor that production—so as our client gets those 1,000 parts and provides feedback—if we haven’t gone through inspecting and going on that journey with them, that feedback doesn’t mean anything. If we’re going to start dialing that in, we have to start over: let’s do a sample run, let’s do a dimensional inspection, so we can become highly familiar with the parts that you need.

Speaker:
But we’ve been doing additive manufacturing for a long time. Aren’t there tolerances? Can’t we just apply standard tolerances? Why do we have to go through this dimensional process?

Speaker:
As a manufacturer who’s making parts for others, you’ve got to become aware of what the actual manufacturing process is that you’re using. What’s the machine that you’re running on? The PPAP process allows that awareness to occur. Then we can align with the customer and make sure that they may be calling for a nominal dimension that’s plus or minus, or -0 plus ten-thousandths, for instance. But it’s not like CNC machining where we can adjust the machine for that one part to hit that non-symmetrical tolerance. It’s geometry-based, and it’s build-orientation-based. There are lots of variables specific to additive that other manufacturing processes don’t experience.

Speaker:
So not only do we have to understand alignment with the project and client expectations, but we also have to understand the results of the printing method with those particular geometries. It’s meshing all of those things, and that’s done through the PPAP process.

Speaker:
Yeah, and the important thing for us is, as soon as our team is aware that we’re doing something that’s going on the journey for a PPAP, we’re going towards production, we’re running samples, we’re doing dimensional controls. We look at those parts differently, because we expect we’re going to see this part, maybe in different versions, in the future. We need to make sure we minimize the variation that we build into the processing so that we’re reducing possible end-part variation at the end of the day. Are we orienting the parts the same way every time, even through rev changes? Are we managing where they’re located in the build the same way? Are we packing those builds in similar approaches, so the variables we control are minimized as much as possible, and we see the least amount of variation on the output?

Speaker:
It makes me think of some of the programs we’ve worked on where we’ve either developed the product in-house and manufactured it ourselves, or we’ve co-developed a product with a client. The amount of iteration and collaboration—you mentioned collaboration early on in this discussion—was important to dial in both the design and the manufacturing process to achieve the desired results. When we’re intentional about that and do it on the frontend, the backend is essentially seamless.

Speaker:
And we know what questions to ask. So as our team sees that coming through, we’re already aware of the potential risks, so we’re asking those questions. When we’re delivering those parts, we know what the risks could be, so we can highlight those to the customer, be as transparent as possible, and make sure we’re accelerating through this phase without too many rework or restart operations.

Speaker:
That’s one option. The other option is some programs, whether it’s perception or reality, don’t have that time available on the frontend. It’s like, we’ve got the design finished and we just have to produce 500 of these parts. That’s a scenario where you skip the PPAP process in a traditional sense and just start producing parts, then you measure results on the backend. In some cases you may have high scrap or high yield loss, and it’s learning in the heat of battle rather than doing that proactively.

Speaker:
Yeah, so each of our customers views resources differently. You’ve got time, you’ve got cash—

Speaker:
Time duration and cash seem to be the two biggest and often competing, where one sometimes wins over the other. Those become hyper-critical to how we manage the PPAP. If we have a hard deadline, like a clinical trial that we have to have functional parts for, and we’re also making revisions from the previous clinical trial, we’ve got a window to operate within. To not do a PPAP there at all is sort of crazy because there’s no chance to redo it. We may have to highly accelerate a PPAP to verify that we’re going to get the results we expect, or acceptable results. In those cases, we have to be highly aligned with our customers so when we take that risk with them, we’re all aware of the potential risk, and it’s the greatest chance of success to hit the deadline.

Speaker:
Often, in other cases, the deadline isn’t so tight. It can extend, worst case, where maybe cash isn’t a problem and we’re going to gamble, and our clients are willing to hedge their bets to have a chance at success with a high-volume production—knowing that some of those parts may not be usable, they may need to reproduce. But they’ll take the learning opportunity, make the adjustments, and maybe the deadline for them changes a little bit, but they’re willing to flex there. They just need to make as much progress as possible.

Speaker:
It’s interesting, as you share those examples, thinking about a program we’re supporting in real time today where the resource limitation is design-time resources. We’re working on a program that is coming up to a clinical build. There isn’t time to do the PPAP, but we have a reasonable assumption of what the results are going to be based on the part design. It’s something like doing a six-week production in a three-week period. Since we haven’t done the PPAP, we’re overproducing the critical parts, a certain percentage we’ve assumed on the frontend, so we can manage risk in terms of time. That’s a case where dollars aren’t as critical as days. We’re basically managing risk by overproducing, which we can do in that timeframe, and then have extra parts we may not use, but at least we have enough to fulfill the needs. We haven’t done the PPAP, we don’t know exactly what the results will be, and we just have to overproduce versus going into it with full knowledge of the results.

Speaker:
That scenario is even more complex because we’ve been producing similar parts in the past for several years now, so we’re highly familiar with the parent part in the assembly. It’s just a newer revision, we’re going to be doing a little bit different operation to it, but we’re building on that context. So the risk is managed the best we can, the client’s aware of it, and this is the best thing we can provide given the outcome. If this was a brand-new client and the first time we interacted with them, the risk would be completely different—there’d be much greater risk.

Speaker:
You said earlier it’d be crazy, and if money was no object, it’s probably not crazy.

Speaker:
It’s not crazy, yeah.

Speaker:
But if money’s an object, then it is crazy. If money and time are both objects, then it’s really crazy. But in some cases there’s just no other option. It reminds me of the saying, “If you don’t have time to do it right the first time, how are you going to have time to do it right the second time?” We run into examples where there’s a perception there isn’t enough time on the frontend, but ultimately there’s time created because it’s done again—the parts are produced again—because the work wasn’t done on the frontend. So when it comes down to it, our urge is to build that time in on the frontend with a PPAP so you can reduce risk and the commotion on the backend.

Speaker:
An easy way to look at that is if you’ve got a large-scale assembly, you’re making 50 or 100 assemblies, highly complex, lots of parts—maybe that total project is $100,000 or $200,000 for all those assemblies delivered. If even 25% of those parts fail and need reproducing, you’ve got $25,000 in initial loss and then reproductions of another $20,000, so you have $50,000 that could be a net loss for that total rework. If you said, “Let’s spend $50,000 on the frontend”—worst case is $25,000 invested on the PPAP, highly accelerated small-part production—let’s get our hands wrapped around it and then go to it. Even if you go through that stage and you’re unaware of the risks, you come out the same at the end.

Speaker:
What’s wild about what you just said—that would be the most extreme and probably not even real scenario. It’s more like if you invested $500 or $5,000 on the frontend for a PPAP, your return on investment is 10 to 100x.

Speaker:
Yeah. Which is massive. The first initial production batch that’s wasted and the second rush—it’s a massive return on investment. This is something we push for regularly. Sometimes it’s painful and sometimes it creates friction, but it’s always done with the best interest of saving that amount you said, preventing rework and preventing the time loss because you’re redoing things. Those are things that we’ve experienced multiple times. In the rush of doing it the first time too fast, you end up doing it again anyway. So we’re always trying to push back on, “Let’s make sure we’re doing this methodically rather than just as fast as possible.”

Speaker:
You know, the PPAP process may not always weed out 100% of everything we’ll experience, especially if we’re going to large-scale production. There are always lessons learned as we go through the PPAP process. If we’re going to produce 5,000 parts or 10,000 parts a year for something, and that PPAP process—even if it includes 500 parts—as we scale up and go into ongoing production, sometimes we have to dial something in. But it tends to be less critical at that point, where it’s not a showstopper; it’s an optimization effort in that production cycle, and it makes everyone’s lives easier. It’s not a complete rebuild, it’s a minor tweak or minor change.

Speaker:
So is every PPAP process exactly the same across all manufacturing, all additive manufacturing methods?

Speaker:
Well, we try to correlate the right style of PPAP for the manufacturing process, the volumes we intend to produce, and how dimensionally critical the part is. We don’t want to overdo it. If we’re making a part that’s more ergonomic and the dimensional controls aren’t so important, but the surface precision is—surface smoothness or avoiding roughness—then that PPAP is tailored to that program. We’ll take things out or keep the bare minimum to get the PPAP aligned. On the other side, if it’s high volume and highly dimensionally critical, we’ll hone in on the areas that are really important, or we’ll discover and then hone in depending on the phase we’re at.

Speaker:
So it really is about understanding risk at the first step. What’s the scope of the program? What’s the manufacturing method? How many parts are going to be produced? Then you manage that risk on the other side, building the program to support the scale and scope of that initiative.

Speaker:
Yeah. My hope is that when a client, new or existing, calls us at DI Labs and talks to one of our team members, they’re hearing things like, “What are you going to use this for? What solution are you trying to solve with this part? What have you experienced in the past? Where do you expect to go with this part or the program in the future?” We can take that knowledge and be most effective with how we produce, making suggestions and getting down this journey of getting this thing to production.

Speaker:
And we’ve got a PPAP process on our website, and this is what we use as the foundation for these programs. Then we shrink it or build it out from there based on the complexities and scope of the program. We also live and have lived lives of product development and leading programs using other manufacturing methods. You shared an example with me this morning of a program you worked on. It’d be interesting to talk a bit about that.

Speaker:
In a past life, I was designing a lot of sheet-metal parts and having them laser cut, and there are a few holes that need to be precise for the project we’re working on. As I was sending in parts, I wasn’t having a whole lot of collaboration with our vendor, but I’d send those parts in, get them back, and measure them. With this feature I had, the bearing wasn’t fitting into the hole. I recognized there was an offset through all the parts across multiple orders— the way they were laser cutting was producing a hole too small. So I began modifying my geometries to account for that offset. Over time, something happened: their machine was recalibrated, and now the new parts I was getting were way oversized. That can happen in nearly any manufacturing process. It was mind-blowing because I said, “I’ve been offsetting dimensions to get good outputs for me. I wasn’t specifying them to be checked by your team, I was just using the results. Let me modify the design to get what I want.” And they’re like, “If you see that in the future, we need to connect on it because they should be coming out accurately.” They had to calibrate the machines, now the dimensions are different, and it took a rework of everything to get back to normal. If that happens with one vendor on one machine, then across manufacturers in the U.S. and globally, machine calibration, environment controls, all these things have a big impact. Without close collaboration, you’re just hopeful you’ll get what you want.

Speaker:
What would you do if you could do that over again?

Speaker:
If I didn’t have to learn that experience again, and I would’ve had that loop happen after I experienced it, I would’ve been more aware of the dimensions I needed, I would’ve specified them. If I got results back on parts that were different than expected, I’d reach out and say, “This isn’t what I was expecting. Is there something going on your end? Can we root-cause analysis this with your team?”

Speaker:
So you could’ve saved time and probably cost by communicating on the frontend, saying, “Hey, this isn’t what I expected. Can we dial this in?”

Speaker:
Yeah.

Speaker:
To me, there are two lessons. One, any manufacturing method is going to have some degree of variability. Two, without a feedback loop, a bidirectional feedback loop where you’re setting expectations, understanding expectations, and then driving results to those, if you converge on a solution, that’s going to be luck—it’s not intentional. So it’s about making sure the manufacturer you’re working with, regardless of process, has clear communication on what’s most critical so the manufacturer can dial those things in—so we at DI Labs can dial those things in—and then measure against them and drive to a converging solution rather than just getting whatever comes out.

Speaker:
Yeah.

Speaker:
It’s also interesting thinking about that calibration fact you shared, because that’s another piece of the internal puzzle we face in the industry as manufacturers. We’re running different pieces of equipment that have different baselines. You can put the same exact part on different machines, especially across different methods, and get different results. Which again is why communication is critical, especially for customers working with multiple vendors. You can get different results, and according to the manufacturer, those different results are all within spec because of the method used. So that communication is critical. For those of you using additive manufacturing—whether you’re using it in-house or not—make sure you’re communicating what the critical dimensions are, what geometry and features are functionally critical. That’s important to communicate back to the manufacturing team so there’s alignment.

Thank you for sticking with us through this podcast. This is just a high-level overview and it’s not intended to capture all the nuances and specificities of each potential geometry and program. We’re here as a resource to help answer questions and guide you through your production activities successfully.

Speaker:
I think the important thing is that we really see ourselves as guides on the journey, and we try to point out the obvious pitfalls. We want to help our clients avoid those.