Two Products, Different Projects
I started building musical instruments last summer. It’s a great way to push the envelope, a challenging and entertaining hobby, and you get something at the end of the day that you can continue to use over the years (if things work out, that is). While there are many things that make each instrument an interesting project, it is not safe to say that every instrument created is a project unto itself.
For someone that has limited experience working with wood, has a few tools and limited space to work in, every aspect of building a new instrument can be a challenge. I need to decide what kind of instrument to build, where I am going to get reference information, source the wood and other supplies, and figure out almost every step of the way how I’m going to perform the next step. There are setbacks along the way, some that I can recover from quickly, some that simply get buried in the remainder of the build, and some that can be expensive and time consuming to recover from.
I start each instrument with a clear understanding of what I want to achieve, and that clarity is becoming more precise with each instrument I tackle: it goes well beyond “I want to complete the project and not have it explode when I string it up”. There is a great deal of learning along the way, and I am careful to log the effort and cost at every step as well as the decisions, the opinions and the learning that points me down one particular path rather than another. At the end of the project – when I have a playable instrument – I reflect back on those goals and check to see how well I have succeeded. It’s not a boolean pass/fail – there are some elements where I exceed expectations, and others where I learn that I need to pull my socks up next time.
All that stated, I don’t build a gantt chart for the project, it’s a project where that view doesn’t provide me the insight I need to progress. I’m not wrangling a number of different people that need to collaborate, I’m not working to a particular deadline, there are no formal milestones for me to achieve. At each point along the way, though, there are usually a dozen or so next steps I can take that will get me toward my overall goal. Some can be knocked off in a few minutes, some just require a little noodling with paper and spreadsheet, some require careful setup and a few hours of uninterrupted time to get the step done. Some are pretty simple, some will require me to ponder for quite some time about how to solve that particular problem. When I’ve got some time to work on it, my current situation and demeanour help determine what I’m going to do at that moment.
Each instrument is clearly a project: it has a start and an end, there are a whack of interrelated decisions to be made along the way, and there is a product that didn’t exist when I’m done.
Compare that to what happens when a major guitar manufacturer spits a new guitar out the end of their assembly line.
With a good quality manufacturer, when you buy a certain model of instrument, it is a pretty safe bet that what you will get is within pretty narrow margins of variability. While the particular grain on the instrument will be subject to the tree that the wood came from, there is not much more that differs from one to the next. Big manufacturers will produce tens or hundreds of instruments a day, and there is very little discovery and uncertainty in doing so.
To do that consistently, to be able to stay in business and make a profit, each instrument cannot be treated like a unique project.
There are projects at those shops, to be sure, but the projects are geared to the development of a process that will create an ongoing line of instruments with particular characteristics. All those design decisions I have when I build a new instrument are squeezed out of the production process along the way.
Craftsmanship is replaced with CNC machines that can fabricate exactly the same neck contour over and over. Each step along the way is analyzed and optimized to reduce cost and time and variability. Specific tools are designed and built so that the instruments can be assembled exactly the same way from one to the next. The people involved are not deeply dependent on one another to collaborate and solve problems: Alice builds parts and they get stashed in a bin for Bob to use as he needs them. While everyone involved may share a passion for music, there is no need for them to have a deep understanding of the entire production process. Lean techniques to manage the supply chain can be a useful way of optimizing the production of instruments, but each one is not a project.
Getting serial number 125 off of the line is not any more difficult than getting the next one off the line, or any easier than the previous one. All the complexity has been squeezed out of the process, done so in the project of designing the process and deciding on the characteristics of this particular model.
In both cases a beautiful instrument can be built: either a single one or a series of clones that look and behave in the same manner.
In your field, look for opportunities to design out all the uncertainty in things you will be doing often. If your product has a complex installation process, you may be better off architecting installability into the system so that this becomes a predictable and efficient process. If you will be porting to multiple platforms, explicitly consider how to ensure this is done with no surprises as you go from one to the next. Working these things out consciously, in a proactive fashion, will pay dividends downstream. It will be more like taking a number of beautiful instruments off an assembly line than the challenges of building a completely new instrument each time.
Both can be interesting, entertaining and rewarding, only one will give you the optimizations you need to make it a profitable business. – JB