Hacker News Reader

Comment by jcfrei | Hacker News Reader

I would argue that the main challenge is actually manufacturing. Building one or a hundred great cars is easy. Building hundreds of thousands great cars with millions of customized variations (like BMW, Audi, etc do.) is the difficult part. Tesla has been behind its promised number of deliveries in 2014 (http://www.forbes.com/sites/greatspeculations/2015/02/12/ear...). If Tesla wants to be a big player in the automotive industry they really need to keep optimizing their supply chain and manufacturing. The product itself is top-notch, no question about it.

View on HN

That's true. Also, most consumers don't realize that the reason why Tim Cook become Apple CEO and why Apple is doing so well financially is their finely tuned supply/manufacturing chain. At high volumes that can make or break a company.

For example, VW will predictably do really well in the next 10-15 years as they've invested tons of money into optimizing their manufacturing; they use the same basic modules for a large VW and a small Seat.

Yup, their MQB platform is pretty cool: http://en.m.wikipedia.org/wiki/Volkswagen_Group_MQB_platform

Can you explain why millions of customized variations are required?

Why not go the Apple approach - provide a small number of models that satisfy a pareto-split percentage (80%+ including the more profitable markets) and leave the remaining percentage either to aftermarket partners or competitors?

That's the approach Tesla has taken, focusing primarily on the highest margin portion, and filtering down.

Tesla isn't planning to be a "big player in the automotive industry". They plan on creating new product market, which just so happens to be compatible with the existing infrastructure. The incompatibilities (i.e., no ICE/gas) is a feature.

I think they'd need at least three different products: a sedan, a minivan, and a truck. To the other commentator, I don't believe market segmentation is essential at this stage of the game yet. However, my information about it is mostly from this Joel Spolksy article http://www.joelonsoftware.com/articles/CamelsandRubberDuckie... so I don't know much about this topic.

Market share for sedans and minivans has been declining for years. Current mainstream USA market growth is mainly in small to medium crossovers (i.e. tall station wagons that sort of look like SUVs).

They are already mostly done with the estate/SUV Model X. Then, they are going to introduce the Model 3, which will be more like an ordinary family car. That will be three models.

> Can you explain why millions of customized variations are required?

Usually that's what thing that differentiates >=premium cars from the rest. If you buy a Skoda, you get packages. If you buy an Audi (same conglomerate), you can customize everything as you like.

Because while people might be OK having the same $1000 iPhone 6 as everyone else, they're not OK having the same exact $80,000 car. They want an extra feature ere or there that the next guy with a BMW m5 doesn't have. They might want some extra carbon bits here or there. Cars are as much fashion as watches, and car customization is a huge industry / widespread interest. If dealers / mfgs can capture even a bit of the aftermarket industry (think simple stuff like chrome exhaust tips / different wheels), it's probably a win.

> Can you explain why millions of customized variations are required?

For the same reason that car manufacturers offer different models for the same product (personal transportation): like the A, C, E, S class (in the case of Mercedes). These classes allow a producer to differentiate between consumers with different spending power. If you offer just one product at one price you can't capture the whole market (ie. you sell too little / too cheap to rich people and can't sell to poor people).

Apple recently started going this way as well with the introduction of the Apple watch. If you look at all the models and options they can cover a very broad price range. Differentiating between the watch sport, the watch and the watch edition. Add to this the different sizes (38MM and 42MM) and the different bands and you get a very nice price continuum between 349$ and 17000$.

And yet, the key tooling (chipset/OS/form factor) for those numerous combinations is a handful of choices in total.

Add in the fact that it's not millions of variations (really only a couple dozen now). Tim Cook famously said in some conference a while back that all of what Apple sells can be fit in on a table in front of him. I don't think that guiding vision has really changed.

Furthermore, it's not clear if the Apple watch will be successful. I wouldn't use it as a guideline for a successful product strategy - yet.

And imagine the deliveries increased an order of magnitude when they try to sell the model 3.

> they really need to keep optimizing their supply chain and manufacturing

And work on more charging stations and battery life. Here at the office the guy with the Tesla is constantly mocked cause he couldn't even make it to the team building event, less than a 3 hr drive away (mainly by the guy who had to borrow his mom's Clio to buy some loudspeakers that didn't fit in his own sports car, but that's a different story ...)

Why on earth you'd spend a hundred grand on a car that can't drive more than a couple hours is beyond me, even if it has a big-ass iPad inside.

You realize this is such a small part of the overall driving experience? I don't know how people are convinced this is a huge deal. It is possible to rent a car for the 1 or 2 days a year you have to drive 400 miles in a single trip. In fact, if you own an M5 I would also suggest you do it as well.

I've had my car in Austin for a year and a half now and never once been concerned about power, never once been to a charging station (and never been to a gas station). I would suggest that most people aren't regularly taking giant road trips.

The touch screen is not even that awesome (no tactile response), it's less than 1/100th of what makes the car great to own / drive.

The problem I have with this argument is that is doesn't eliminate the problem, but merely removes it a step. These are problem EVs need to surmount because other wise we will have lots of of personally owned EVs that get parked so people can rent ICE vehicles every time they leave town for a trip. As an early adopter people expect to have to look past these issues, but for wide mainstream adoption, and dare I say, replacement of ICE vehicles, the problem will need to be solved with a combination of better range, quicker changing, and a wider availability of quick chargers.

Also, there are many people who take long road trips far more frequently and for whom range is a genuine issue (between my vehicles, I clock about 30k miles a year--mostly pleasure as I live only a mile from my job.) I don't begrudge those who can make an EV work for them, but it won't for me.

> Also, there are many people who take long road trips far more frequently and for whom range is a genuine issue

Yes, that is true. People will need different ways of commuting based on what their needs are.

> The problem I have with this argument is that is doesn't eliminate the problem, but merely removes it a step.

Also true, but what is the problem? Again, you need different ways of commuting based on what your needs are.

No one suggests taking cars everywhere. Some people like road trips, but many people fly when the distance is too far and their time too short.

Likewise people in New York take the subway, elsewhere people ride their bikes. No one suggests the fatal flaw with those is that you can't take them to the next state over. You just rent a car or take a bus, or do whatever, or, if you really need it, you pay a decent sum of money (in New York, at least) to keep a car in a parking space for when it is needed.

We're already in a no single size fits all situation, so it's not clear what the problem with that continuing is.

Yes, similarly, nobody says an SUV that get 10 highway MPG are bad for road trips even though clocking 30,000 miles per year will cost an extra 3 thousand (and that's after the fall in gas prices) dollars per year over a 20mpg vehicle (which is still low).

After depreciation and gas, might be cheaper to rent a car for them too.

If my Google car is a taxi, why do I want/care about custom variations?

Reply on news.ycombinator.com