A Bit About the Production Delays

This post is regarding the backlog and the models related to that, which would be the first runs of newer spec GSO-5.1s, GSO-4.7s, GSO-2.7s, GSO-4.1s, and GSO-7/7s. First off, a little context.

The Starter campaign provided us with an initial boost, to purchase new equipment and expand our production run sizes. We purchased the finishing equipment, materials, got ourselves into a bigger shop to handle the space/power needs, and plotted a solid path toward expanding production. We hired a promising young guy to help in the shop, someone who had worked for us back East. We also hired a talented young woman to help out with social media and customer service. We had companies with the heavy machining capabilities lined up and ready to help us grow. Everything was looking great... Until the first parts began arriving.

The GSO-5.1 blades had uneven bevels, heavy surface grinding lines and the blade profiles weren't quite matching up with the handles. It wasn't an ideal start but also not a huge deal, yet. We worked through it and talked to our service providers about what could be changed to improve things moving forward. That run took a good bit longer than we had expected but as long as the issues were resolved moving forward, everything was going to be just fine.

Then the GSO-4.7 blades arrived and we were still having the same issues. Some time had passed by this point and folks were getting anxious for progress. "I mean they're just production knives, you're basically just slapping handles on blades and sending them out."

Side note, to those who keep belittling our work by saying "it's just a production knife, why does it take so long?!", like they're coming out of a photocopier or something, you are grossly oversimplifying the process. There are a lot of moving parts when turning a sheet of steel into a finished knife. People are behind every step and manufacturing large batches of knives to a very high standard takes a lot of hard work. Each blade blank is precision perimeter machined and that requires constant attention on the part of the operator. Maintaining a consistent tolerance over many hundreds of parts and ensuring the finish quality isn't suffering as the tooling wears down is a constant battle. Machining CPM steels isn't like machining aluminum, it is hard on tooling and even the machines themselves.

When surface grinding the blades, a human being is doing that. Careful attention must be paid to avoid overheating and warping the blades on the table, removing the right amount of material, and achieving a high enough level of finish that it doesn't show through in a completed knife. Similarly, bevel grinding is a constant struggle. It is expected that plunges, bevels, and breakouts at the tip are all going to match from side to side. That stuff doesn't just happen and the operator really needs to be on their shit. The bevel grinding process falls somewhere between machining and witchcraft to get everything right. Eventually we'll be able to get a process established that is a bit more predictable but that is a $500k+ investment. Not a casual purchase to say the least. That said, I feel confident we'll eventually get there.

Now the blades are done, it is time to start machining handles. We're looking for as close to a flush fit with the blades as possible, with no overhang of the handles. To achieve this we're shooting for tolerances of flush to -.002" on finished assemblies, which doesn't leave us very much room tolerance wise. Usually we’re shooting for the number on the blueprint +/-.0005", for about .001" total tolerance for the blades and then the handles. It is honestly hard to reliably measure tolerances that small with standard measuring tools but we're doing our best. Special attention must also be paid to the handle hole tolerance, so the fasteners don't fit too tight and the heads aren't too proud or too deep. Again, you're fighting tool wear, so constant adjustment is required to keep everything consistent and as close to perfect as possible. Even the machines themselves have a maximum tolerance they're capable of, which needs to be accounted for. That can change based on the temperature of the room, the build quality of a particular machine and the age of the unit. Down here in Meridian I push the guys for absolute perfection, so when something is a little bit off, it truly is just a little bit off. An aim small, shoot small approach. I drive them absolutely crazy with my demands but we’re all happy in the end, when everything comes together and the final product is beautiful.

Even something as seemingly simple as our screws, precision machined from bar stock on a swiss turn lathe, need constant attention to ensure everything is right. A great recent example, we just needed to have 3,000+ components remade because the threads got out of tolerance while the machine was running lights out. They look fantastic but won't thread together at all.

I guess the point of my little rant is that it’s always something with manufacturing, especially at scale. Things go wrong constantly even with me in the shop every day, keeping a very close eye on my projects. Now imagine trying to subcontract these services to someone in a different state. Things get lost in translation, details are overlooked, expectations aren't properly conveyed to operators, your expectations are not aligned with what your service provider is actually willing or able to accomplish, project managers misunderstand which aspects of a part or assembly are critical to you, an inexperienced employee runs your parts, an operator shows up hungover or has some personal stuff going on and isn't paying attention... You get the picture. When you’re a demanding customer, an offsite service provider is going to have a hard time meeting your expectations.

Ok, back to the story. Despite my better judgement, we kept pushing more models through production to keep showing progress, again with promises that the work was going to be constantly improving. As run after run showed up, all needing a lot of individual attention, the delays became compounded. We were starting to look at what other processes we could bring in house to help deal with the issues but then the internet happened.

Every internet troll, self-proclaimed forum expert, blogger, youtuber, competitor, and random person with a keyboard decided it was time to jump online to either share their opinion on the matter or attempt to incite an online riot aimed at us. We were suddenly scammers, liars, or perhaps just the ringleaders of a giant ponzi scheme who weren’t really making anything at all. People started sending us wildly inappropriate emails and even death threats. Something you should know, something I had to learn through painful experience, is that if you decide to do anything that requires you to interact with others on the internet, be prepared to grow really thick skin. To the rest of the world you’re no longer a human being, you’re just more internet content and will be treated accordingly. People will post the most awful messages and comments imaginable, things they would never even consider saying to another human being in real life, all without a second thought or regret. 

Anyway, the timing couldn’t have been worse. A flood of refund requests came pouring in, at a time when we needed cashflow the most to help get on top of the situation. Despite what some people seem to think, the money isn’t endless. You can’t always just “do” things. We don’t have the margins of an imported knife and at the time we didn’t have all of the in house resources of a full manufacturing company. When this whole thing kicked off, we had planned on things going a certain way, in a certain timeframe. We were heavily invested in equipment and materials. We had monthly fixed costs like payroll, taxes, rent and utilities on the shop as well as our personal rent and groceries. There was also all of the money we had paid to our service providers for things we weren’t yet able to sell. All of this severely limited our options. At the time I only saw two real paths forward. Keep pressing on the best I could, with the means I had available to me or call it quits. With how things were going and apparently the entire internet rooting for our failure, I admittedly went to a pretty dark place for a bit. In the end though I knew quitting wasn’t an option. In spite of how horribly wrong everything had gone, I had made promises. People out there still believed in SURVIVE! and were counting on me.

So we got as skinny as we could and for quite a while stopped taking orders entirely while working through our backlog. With help from our families, friends, and some great customers, we eventually saw our way through the worst of it. By meeting Millit and moving to Meridian we were able to reduce our overhead even further by renting just a corner of their shop for a bit. The addition of our finishing equipment also helped fill some gaps in their capabilities. Their equipment helped us get through our backlog even faster by using cnc machines to fix handle perimeters and by using their larger table surface grinder to refinish substandard surface grinding.

To the question of why we don’t just throw all of our resources into finishing out GSO-7/7 orders more quickly, the simple answer is there just isn’t any money in them to justify doing that. We’ve already paid for all of the work once, even if that work wasn’t up my standard. We paid Millit to resurface grind a lot of blades to save ourselves some time but even that was a slow and expensive proposition. Employees don’t work for free, so most of their efforts are being focused on the upcoming, more profitable endeavors. I’m the only shop guy I don’t need to pay, I have the best idea of what is going on with this older work, so I’m the one stuck doing it. It isn’t the fastest approach to getting these oldest orders delivered but it is the approach that gets us out of this mess and on to better things moving forward. We’re still delivering knives in the order they were placed, despite all of the other stuff going on. We have lots of knives moving through production now as we ramp up to take over most of Millit’s capacity as some of their other projects are wrapping up.

Progress is being made and each day things are getting better. We’re finally back in a place where we’re confident enough to invest (through personal loans) in more equipment to further improve production output, and we’re very much looking forward to a brighter tomorrow. Check back soon for our next post outlining how we plan to fulfill all existing preorders by the end of 2020.