Framebuilder Survey Analysis: Business Demographics

I’ll turn next to what I think of as the business demographics of framebuilders – recognizing that these category headings don’t have super distinct boundaries. Here I’ll focus on different aspects of framebuilders’ businesses, leaving more detailed analysis of production and output types/categories, along with consideration of selling prices and such, to later posts. Tables below are rough-and-ready dumped from Excel (I hope the final version will be prettier, but these will have to do for now given the busy-ness of the times).

But First: A Big Caveat/Reminder

The sample I am working from in these analyses intentionally excludes larger-output shops that are handbuilding bikes in the U.S.! As such, this is not representative of the population “all bicycle building shops in the U.S.” – nor was it intended to be!

Why does the sample exclude larger shops? Aren’t they important?

Yes, they are important, and I intend to incorporate information from those larger, typically more production/stock-oriented shops (for instance, places like Co-Motion, Moots, Waterford/Gunnar, etc.) in future analyses. And, this would really be a problem if I were trying to estimate the total output of the U.S. handbuilt segment from this sample (an exercise I engaged in more hypothetically a good while back on this blog).

However, I excluded larger shops for a few reasons.

  • The types questions that make sense, or can be applied, to small-scale builders don’t make sense with larger shops. Questions about individual motivations, orientations and practices don’t really work for someone managing a multi-employee production facility that they themselves do not own or directly control. Likewise, salary and household income data don’t apply if someone is replying on behalf of a larger firm.
  • The theoretical/analytical focus of this project has been to understand craft/artisan fabricators and producers and how they make a livelihood within the market, so this also justifies my emphasis on small-scale producers…rather than “all bicycle builders in the U.S.”
  • Large shops were also excluded because they are a numerical minority of framebuilding operations overall. Put differently, while bigger shops might make a big contribution to the total output of bikes fabricated in the U.S….the vast majority of builders in the U.S. are small-scale (often single-person) shops.
  • Finally, I did have a few larger shops respond to my survey, but I’ve excluded them from the analyses (for the time being) because they would so heavily skew certain results. Having a couple of shops with an output of, say, 300-400 units a year in my sample would bump the “mean output of bikes per year” up enormously, for instance. Now, I’ll be first to admit that the distinction between a “small” and “large” or “mid-size” bicycle producer is fundamentally arbitrary, so I can’t make a strong argument  for a precise cut-off point between large and small. It’s a bit of the “you know it when you see it” definition. In this case, though, I used a combination of total output (excluding those over about 150 units a year), employees (excluding shops with more than 5 employees), and production type (excluding those that were majority stock models). Putting those together, in plainer English, shops that were using lots of employees to building lots of units of standardized bikes each year were excluded (again, for the time being).
    • A nerdier sidebar to that last point. One might reasonably say: if you are pulling a sample of “all bicycle builders in the U.S.”, and if you say that there aren’t that many larger shops in the U.S. (15 or so?), then isn’t the fact that you only got a handful of responses from larger shops still reasonable from a sampling perspective because, after all, you may have received responses from a similar portion of the larger shops as you did from the smaller shops? In other words, wouldn’t you still have a representative sample? I would say that, yes, this would be the case….but only if I had solicited responses from all of the big shops. I did not do this; the responses from larger shops that I did receive probably came from the survey link being open and advertised on multiple sites/podcasts.

OK, so on to the analysis at hand….

How long have you been building bikes overall?

  • Mean: 16 years
  • Median: 12 years

What is the age of your current business?

  • Mean: 14 years
  • Median: 10 years

Do you currently carry liability insurance? 85% answered “Yes”

While this is a sizable proportion, it’s smaller than one might have guessed. However, I think this is explained by a variable I will bring in for more analysis in future posts: whether the builder is “full time” (in their own view) or not. Of those “full time” builders, only 4% report not carrying liability insurance. Interestingly, though, 73% of non-“full time” builders also report carrying insurance!

Does your business have other owners (in addition to the respondent)?

  • Sole owner: 92%
    • And, of the 8% with multiple owners, most of them (78%) have only 2 owners.

Do you have any employees (apart from owners)?: 16% say “Yes”

The distribution of employees by part-time and full-time status looks like this:


Keep in mind that the actual numbers in that table are quite small (with only 17 builders reporting have any employees!). In fact, the numbers of builders in those different categories are so low that I’ve not provided the raw counts because I worry it might allow readers to identify specific builders!

Where are your customers located?:

To what extent do you rely on the following sales channels?:

So what does it all mean?

Keeping in mind that I’ve excluded large-output facilities (of which there aren’t very many in the U.S.), these are primarily single-person shops (“one man shops” if you look back at the gender distribution from the builder demographics!) with few employees and few business partners. Most of these builders are selling almost entirely direct to consumer, with some selling at least partially through dealers. As before, no big surprises here – but, as before, that helps to confirm that my sample seems pretty representative, just thinking about the stereotype of the “typical” U.S. framebuilder that most people carry around in their minds.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s