Tesla has been in hell for the last year. Production hell, that is. The brutally long weeks and restless nights spent sleeping on the factory floor had taken a toll on Elon Musk and the team at Tesla, seemingly pushing him near the breaking point.
On the Q1 2018 investor conference call a few months back, Elon cracked, interrupting an analyst mid question, calling the question boring, doing something similar with another analyst, and bullishly pushing past the queue of analysts representing institutional investors in order to give the rest of the question and answer time on the call to a retail investor who moonlights as the host of a successful YouTube channel called Hyperchange. Hyperchange is dedicated to documenting “the current economic era of perpetually accelerating disruption.” Thankfully, Gali over at his channel Hyperchange absolutely crushed it with his line of questions and added a ton of value to the call.
Back at Tesla, Elon was still in the throes of misery, wondering if this production hell had become his new base reality. As Winston Churchill famously said, “If you’re going through hell, keep going.” Thankfully, Elon and the team at Tesla kept on going, and after flying a new production line in from Germany, standing it up in a new tent in the north yard of Tesla’s Fremont factory, and cannibalizing some of Tesla’s Energy products for their battery cells, Tesla was able to hit its target of 5,000 Model 3s in a single week.
At the time, many believed it wasn’t sustainable. How can building cars in a tent be the way a car company builds cars? Jerome Guillen, Tesla’s former service lead who now heads up Tesla’s semi truck program, chimed in about the tent to share that, “It’s permanent, for now.”
That one line speaks to Tesla’s modus operandi for its entire business. It develops new Autopilot hardware that is the best on the market … then iterates.
We learned on the second quarter earnings call yesterday that the Model 3 Autopilot computer was designed to be replaceable and that they have been working on a replacement for that very computer that’s 10x faster than the world’s previous state of the art option! This car just moved into production ~14 months ago and they already have a new computer that’s 10x better. That’s nuts. That’s Tesla.
Getting back to the Model 3, this car has taken Elon and the team on yet another roller-coaster ride. Few expected it to last this long. The shorts thought Elon’s head would explode, leaving the company bankwupt on the side of the Bay, selling off its assets to pay its creditors. The longs rode the roller coaster of trials and tribulations, but many wondered almost daily if their position was real or if they’d be left holding little more than an empty account statement if the increasingly aggressive shorts actually were right or somehow forced Tesla into a deep ditch.
However, Elon shared with investors yesterday that after achieving their long sought after Model 3 production rate of 5,000 cars per week in June, the company has refocused on fine tuning the two general assembly lines. General assembly is where Tesla combines all of the components of the car — bolted onto the painted metal vehicle frame — to create the vehicle you drive into your garage.
Elon shared that they are confident that each of these two lines can support a throughput rate of 5,000 Model 3s per week, but that they were both sub-optimized in their own unique ways. Kicking off what was a trend of non-time-bound statements, Tesla shared that general assembly line #3 — or GA3 for short — is expected to achieve its target production rate of 5,000 Model 3s per week “very soon.”
GA4 was flown in from Germany and assembled in a “tent” in just a couple of weeks to enable Tesla to hit its target production rate of 5,000 cars per week before stabilizing GA3. Long term, that is exactly the name of the game — stability. In July, after a brief respite for the Independence Day holidays, Tesla demonstrated several times that it was able to achieve the 5,000 Model 3 per week production rate, setting it up to achieve a near-term target of 6,000 Model 3s per week by the end of August.
Sustaining its demonstrated 5,000 Model 3s a week production rate is the floor at which Tesla sees itself being sustainably profitable from here on out. “A total vehicle output of 7,000 vehicles per week, or 350,000 per year, should enable Tesla to become sustainably profitable for the first time in our history — and we expect to grow our production rate further in Q3.” That means a near-term push from 5,000 Model 3s per week to 6,000 per week, as well as the longer-term ramp up to its target of 10,000 Model 3s per week at Fremont.
That sounds trivial compared to the massive numbers thrown around on these calls, but stepping back and looking at it over a year, that’s 520,000 Model 3s per year (or 500,000 a year if you take two weeks off). Stacking those cars on top of Tesla’s ongoing base of 2,000 Model S and X per week, you have a very healthy production target for Tesla’s Fremont factory with existing equipment. There will inevitably be some additional CapEx required to fine tune things, but the equipment and the production lines are there. To save you the trip over to excel or your calculator app, that’s 624,000 vehicles per year from Fremont.
Don’t go running off to buy Tesla stock on my account. The production rate of 10,000 Model 3s per week is one Tesla expects to achieve “as soon as we can,” according to Musk. Officially, they’re looking to hit that production rate “sometime next year.” A year from now may sound like an eternity, but just one year ago, Tesla was struggling to make a few hundred Model 3s per week, let alone the thousands it’s making now.
Elon credits this to the continuous improvement culture at Tesla, and how they continue to discover new opportunities to do what they do more efficiently, more rapidly, and more cost effectively. “The potential for our existing lines to produce more cars is much greater than expected,” he shared on the call.
One significant backpedaling they have made is with regards to automation. They spoke about how automation was thought to be the key to building the Model 3 at high volumes and low cost. Don’t be mistaken — highly automated production lines continue to be key in many parts of the factory — including stamping, body-welding, the paint shop, powertrain assembly, and battery pack assembly — but it was not the automatic holy grail of automotive manufacturing Tesla may have thought it to be.
Having said that, it did allow Tesla to figure out where it was a good fit and where it wasn’t. Now it’s time to optimize the system — or continue optimizing the system, as this is what Tesla has been doing for over a year. In a fun exchange, Elon and JB talked about how this optimization is resulting in the automation of additional processes that are still being done manually as well as swapping out some automatons with humans to perform the task manually. This iterative process sits at the core of not just how Tesla builds its cars, but how it does everything. It is first principles thinking meshed together with software development’s Agile Methodology that has turned everyone at Tesla into ninjas, hell bent on hacking together a new process until they get it right … over and over and over again.
These gains in production volume have not come at the expense of the foundational metric that is quality, but rather, they have come with higher quality. Tesla shared that Model 3 quality continues to improve in spite of these massive disruptions to production, which is a small miracle. They shared that Model 3 quality is already on par with their quality targets for Model S and X.
The proof is in the pudding on whether or not Tesla’s iterations are working. Elon opened up the call with fun stats about how the Model 3’s market share in the mid-sized luxury segment exceeded sales of several top models combined.
As it continues to hone Model 3 production lines in Fremont, Tesla expects to produce an astonishing 50,000–55,000 Model 3s in the third quarter of this year (July–September). If it can achieve that, it would succeed in delivering a 75–92% increase in production versus the second quarter, with guidance that deliveries should exceed even those numbers based on the thousands of Tesla’s vehicles that were conveniently in transit to customers at the end of the second quarter.
Given the trials of the last year for Tesla, this conference call was the icing on a cake that has waited far too long to be eaten. The earnings call had a triumphant yet calm and resolute tone to it, as Tesla execs were clearly exhausted from the company’s extended journey through 7 layers of hell but also determined to keep pushing forward in their quest to change the way the people of the world get around, generate power, and store power for generations to come.