La Fraise has published this article on his blog. Interesting reading…
6 Likes
Very interesting. It’s a good thing that the carbon-fibre “GCN bikes” have such high margins on them or handbuilt ones would not be able to be even vaguely competitive!
100-150 hours to build a frame and fork the traditional way sounds about right, especially if it has deep custom features. But it probably takes a tenth of that in a production-line environment like Maxway.
Because it’s a hobby for me, I only charge people for the materials and parts. It’s still quite hard to get under the cost of a bike they could buy in the shops, because OEMs also get such good deals on groupsets.
2 Likes
A good, refreshing and somewhat comforting read. Kudos to La Fraise. I’m now considering putting my own review on my website simply for transparency to my customers 
Earlier last year I became super stressed in every aspect, mentally, financially, etc. Working my ass off, constant pressure and still broke.
So I made a similar review and put my prices up. I knew demand would drop so I went as far as I felt I could while still actually taking orders. The result was worse than I expected, orders fell off a cliff. Operating expenses don’t stop just cus I haven’t got anything to work on. And I know there are others with very similar experience. It’s a really difficult balance.
If I had charged what I really needed to, I simply wouldn’t have sold anything. The business was unviable.
So that’s where I’m at now: figure out how to do it cheaper. It’s that simple. But no easy task because I think I was already working at a high efficiency.
7 Likes
Exactly the same experience Thom. I think we exist in the twilight zone. Too expensive for most and not expensive/exclusive enough for some. I could be a little up myself thinking that too. Have struggled to figure it out to be honest.
4 Likes
Yeah I agree mate. I think it’s even trickier for the mtb builders. The culture is to get the absolute best discount you can, it’s so entrenched that I even get people asking me for a discount. I get it, a mtb isn’t usually the ‘bike for life’ that others can be. It can be hard to justify a higher price tag for something you’re actively destroying a little bit each ride.
2 Likes
I think there’s a missing assumption in this whole discussion that’s worth naming.
The idea that the only way to increase output is to systemise, batch, or redesign the work around efficiency isn’t historically true for framebuilding. It’s a modern conclusion drawn from modern conditions.
Traditionally, lugged frames were not slow to build, they were slow to learn. Even the idea that building traditionally takes 20-40 hours is not accurate.
Once a builder was fluent, cutting, filing, drilling, mitring, and fitting were not discrete operations that needed fixtures, procedures, or protection. They were embodied skills. Judgement lived in the hands. That fluency is why it was once normal for a skilled builder to produce three to five lugged frames a week without batching or production systems.
If builders today cannot reach viable output without systemising the work, that tells us something important has been lost upstream.
It doesn’t mean craft is inherently inefficient.
It means the conditions required to develop fluency no longer exist.
Low volume, broken apprenticeships, early reliance on fixtures, and economic precarity all push people to externalise judgement into systems before it has ever had a chance to form. Once that happens, batching feels inevitable, because transitions are no longer effortless.
That isn’t a moral failure on anyone’s part, but it is a category issue.
There is nothing wrong with running a small scale manufacturing operation, batching work, and designing out variability to survive. That can be honest, skilled, and necessary work. But it is not the same thing as framebuilding as a craft rooted in fluency, lineage, and the workmanship of risk.
The problem comes when both models are called the same thing.
When that happens, new builders are misled about what skills they actually need, slowness gets treated as inevitable, and the craft itself is blamed for economic failure.
If the goal is micro manufacturing, we should say so openly and design for scale.
If the goal is craft, then the uncomfortable truth is that speed has to be earned through repetition, volume, and long training, not engineered through systems.
Until we separate those two paths honestly, we’re going to keep having the same conversation, and people will keep building businesses that fail not because the work has no value, but because the wrong model was assumed from the start.
4 Likes
To be fair traditional lugged frames were kind of all the same though. Same angles, same tube diameters. Same dropouts and rear spacing. The top and down tubes may even have been the same lengths regardless of frame size.
This video below was super-interesting. They also touched on the idea of more “semi-custom” options to make things faster. It probably also makes it easier for customers if you have some at least approximately defined products for them to choose from.
My own experience (as a hobbyist builder) is that I’m surprised at how quickly I can make a basic road or fixed-gear style frame of the kind I’ve made before but how much longer it takes when you have some slightly rad geo and/or an electric motor etc etc.
3 Likes
I’m not sure most of us are looking at building and selling bikes enmasse here. Most of us are trying to attract more customers but are fighting against a market that has been swayed by gimmicks and the promise all the tech will make their riding faster. The right tech can in the right hands but we all know the right fitting appropriate specd bike will always be better.
There is definitely a lack of pathway and progression within the craft. That’s been pushed aside from economics and the very tough retail landscape that frame builders exist in. If you are willing to move and find a position, even go to Asia there are opportunities but they are few and far between.
2 Likes
I have to push back on this, because it simply isn’t true.
Traditional lugged frames were not “all the same”. I can pull frame order books going back forty years, and every single one shows variation in angles, top tube lengths, fork rake, chainstay length, often down to an eighth of an inch. That wasn’t accidental, and it wasn’t crude. It was judgement.
Builders worked within tubing ranges and component standards, yes, but geometry was routinely adjusted to rider size, strength, purpose, and use. Those bikes weren’t generic. Many of them were raced at a very high level and won. They worked because the people who built them knew exactly what they were doing.
What’s happening here is a modern projection. Because variation today is marketed loudly as “innovation”, people assume historical consistency meant lack of thought. In reality it meant restraint. Road and touring riding haven’t fundamentally changed. Human bodies haven’t changed. The bicycle didn’t need reinventing every five years to work well.
Flattening a hundred years of skilled work into “they were all the same” doesn’t just get the history wrong, it does a disservice to the builders who achieved extraordinary results with judgement, experience, and fluency rather than novelty.
You don’t have to like traditional craft, but it should at least be described honestly.
3 Likes
I largely agree with you, especially on gimmicks versus fit and appropriate spec. That’s always been true, and it’s why traditional bikes still make so much sense.
Where I think things have gone wrong is earlier in the chain. The lack of pathway didn’t just remove training, it removed fluency. Without volume, repetition, and long apprenticeships, builders never get to the point where the work lives in the hands. So output stays low, stress stays high, and people redesign the work around systems and novelty just to survive.
That isn’t a moral failure, it’s structural. But when those survival strategies then get presented as “what craft now is”, we end up blaming the craft itself for economic failure, rather than admitting we’ve lost the conditions that once made it viable.
The tragedy is that the bikes haven’t really changed. Riding hasn’t changed. What’s changed is how people are allowed to learn, and how quickly they’re expected to sell.
4 Likes
Cycling and framebuilding have evolved, but their tectonic logic has not. For instance, a 3D-printed cluster is conceptually continuous with a cast lug: both perform the same structural and connective role. What changes is the toolset and the sequence of operations. These tools do not simplify the process; they often increase its complexity and demand new forms of skill.
Claiming that craft has been “lost” assumes it is only about hand control. In reality, craft has always been about tool control. The hand has never worked alone; it has always operated through instruments.
So when someone insists we must “go back” to an older notion of craft, they are missing the point entirely. The skills are still there; they have simply evolved with the process. And yet, every time this argument resurfaces, it plays out in exactly the same way, over and over again, like Groundhog Day.
4 Likes
Claiming that craft has been “lost” assumes it is only about hand control. In reality, craft has always been about tool control. The hand has never worked alone; it has always operated through instruments.
Your comments have valid points. But at the end of the day, if one isn’t trained, or has gone through the equivalent of full-immersion, making a living as a framebuilder will still be a reach. Perhaps I misunderstood the link and story about framebuilding in France. If one doesn’t feel properly remunerated, or can’t find a market, it’s not about the trade (or, er - the craft.) It’s about the man or woman wanting more than he or she has - or expects. Without knowing anyone’s personal story, it’s hard to throw a bone. Whether you want machines, robots, a room full of Sputnik tools, or simply to add some real food to a diet of top ramen, the only way to get it is to practice, and practice more. This, of course, after you’re properly trained for the task. What we have in the Y2K era are a lot of people who want a piece of the framebuilding pie, not unlike others elsewhere want to be coffee roasters, or craft brewers, or bean to bar chocolatiers. One doesn’t just wake up one day and become these makers. But the internet seems to breed these folks, or so it seems from this side of my MacBook Air.
4 Likes
I have to push back on this, because it simply isn’t true.
That’s fine! I appreciate hearing your take (and mine certainly might be wrong 
3 Likes
I agree that practice leads to mastery, but it’s not just about hours spent. Time management and focus on each frame also play a crucial role. I’ve seen people with thousands of work hours produce bicycles that are merely “okay,” while others, with only a handful of frames under their belt, create truly exceptional pieces. Modern society has accelerated the learning process and compressed time: we now have access to far more information and can mix knowledge from many different sources, which challenges the traditional figure of apprenticeship. This has also changed the economics of the craft, as many framebuilders can barely make a living and often cannot afford to take on someone else to train unless they meaningfully contribute to the process. The point behind La Fraise is to understand these differences, price your work accordingly, and ensure you can sustain yourself in the craft.
5 Likes
Agreed. No one has equated time spent with knowledge gained.
From my side, it seems like everyone has his own metric re what’s “good” and some inate indication of what makes a structurally sound frame. I’ve long wondered how one can discern how one thing can be “okay” while another can be considered exceptional WITHOUT seeing the entire fabrication process. As someone whose background is in the sport, I have seen many, many bicycle frames that appear heavy-handed and or seem to represent the work of someone with two left hands, yet these machines rarely seem to hinder the person pedaling it. I’ve also seen a metric ton of beautifully finished frames that, to me at least, have visible design elements (fork rake, clearances, poor rider position) that have left me wondering.
3 Likes
I think this is where we need to be precise about terms.
Judgement isn’t the same thing as knowledge, focus, or time management. Judgement is pattern recognition under uncertainty, and it’s built through volume of exposure to real outcomes. That’s well established in neuroscience. It lives below conscious reasoning.
Access to more information doesn’t compress that process. It gives you better descriptions, but judgement is formed by repeated encounters with success, failure, repair, and consequence.
Someone can absolutely build a beautiful or even exceptional frame early on. That doesn’t mean they’ve accumulated deep craft judgement. You only see that over variation, edge cases, repairs, and time in service, none of which are visible in the finished object alone.
Builders who started young, and worked through the era when most road bikes were craft made, simply built and repaired far more frames than is possible today. That volume matters, whether we like it or not.
This isn’t about who’s “better”. It’s about being honest that some forms of fluency cannot be compressed, only accumulated. We can adapt to that reality, but we can’t wish it away.
2 Likes
This is a great conversation, I really appreciate the perspectives and wisdom.
As a hobby builder it took me about 45 seconds with a calculator to know that for me in this life it wouldn’t be worth the hours of disciplined practice and dollars of investment to make a go as a “real” frame builder. The numbers are just too tight for my reality with a family, health insurance, mortgage, etc. Kudos to all who take the leap and go for it, and to the few who have figured it out and made it work.
I’ve heard rumors that you can count on one hand the number of frame builders in the US that actually make a living just building frames- the rest rely on other income/partners to pay the bills.
4 Likes
I think your last paragraph is way closer to the truth than we want to admit to as a group. Definitely in Australia there are only a couple of builders who stand at the bench and can honestly say they make a living. The rest of us fool ourselves we are doing it when really we are just porpping it up by other means.
3 Likes
I’m also being precise about terms. Richard’s right that we can’t judge craft quality from appearance alone. The relationship between finish and structural integrity, between aesthetics and function, isn’t straightforward. Ellis’s point about judgement as pattern recognition under uncertainty, built through exposure to outcomes over time, speaks to how we learn to make those distinctions. Volume matters. The question is whether the conditions that historically enabled that volume were fundamentally different from what’s available today.
The framebuilders who accumulated that volume were working B2B: supplying a personal brand, shops, teams, and distributors in an era when most road bikes were locally craft-made. That wasn’t fundamentally different from what many of us are doing now to survive. It was B2B production then, and it’s B2B production now. What changed isn’t the production structure, it’s that mass manufacturing captured most of the market, shrinking the space where B2B craft production operates. If volume through B2B with commercial liability was valid then, it’s valid now.
Where I disagree is on treating accumulated experience as the only legitimate path to judgement. When judgement becomes inaccessible to validation, it stops being knowledge and becomes authority. The master–apprentice model didn’t just transmit craft but controlled who got to claim legitimacy. That worked when economics allowed for long apprenticeships that our time can’t afford anymore. Defending it as the only valid path isn’t preserving craft, it’s gatekeeping.
The question, then, is how judgement can be formed when traditional apprenticeship isn’t economically viable. Craft isn’t just about hand skill, it’s about understanding tools, processes, materials, and constraints. Access to better information doesn’t replace judgement, but it does accelerate learning and allows judgement to be examined rather than inherited uncritically. Andrew Denham at TBA and Andrés Arregui at ETB demonstrated this: structured learning that gets people to functional competence faster, then lets them develop deeper judgement through practice. When formalized knowledge transfer is paired with production volume and commercial liability through B2B work, you get both the exposure to consequences Ellis values and the systematic learning that makes that exposure more effective.
This matters because the nature of knowledge accumulation has changed. A builder in 1975 who made 200 frames but never documented outcomes, never registered warranty claims systematically, never formalized failure analysis, just accumulated tacit knowledge that remained inaccessible to examination and transmission. A builder today who produces 50 frames with documented testing, registered materials analysis, formalized iteration, and direct accountability for performance is accumulating judgement that can be examined, transmitted, and built upon. Knowledge that cannot be shared is not knowledge, it’s individual expertise that dies with the practitioner. Volume without documentation keeps knowledge tacit and gatekept. Focused iteration with clear registration of consequences makes knowledge accessible.
Some fluencies can’t be compressed. I’m not disputing that. But if judgement requires exposure to consequences through volume and commercial accountability, then B2B production remains a viable path to keep framebuilding alive. The real question is how we combine that production structure with better learning systems so judgement can be accumulated under contemporary constraints. Romanticizing past conditions as if they were fundamentally different, then using that to gatekeep who can claim legitimate craft knowledge, isn’t preserving tradition. It’s mistaking nostalgia for rigor.
8 Likes
I think this is why the discussion keeps looping. We’re no longer disagreeing about examples, we’re disagreeing about definitions. At the moment, I’m working within an established body of craft theory, Pye, Sennett, and others, which defines craft structurally around judgement, risk, and time. If those premises hold, then certain conclusions follow.
If you believe those premises are wrong or obsolete, that’s a legitimate position, but it isn’t something that can be settled by assertion. It would require a redefinition of craft that engages directly with that literature and explains where it fails. Until that happens, I don’t see how we get out of this Groundhog Day loop, because we’re not arguing on the same terms.
I’m happy to continue the discussion, but either we work within the accepted framework, or we acknowledge that a new one is being proposed and evaluate it properly. Otherwise we’re just talking past each other.
1 Like