Should Tesla publish monthly sales numbers?

[techmaven]

Thanks, everyone who participated in this discussion. It has been an eye opener for me. I had no idea that this was such a sensitive issue. This seems to touch on some fairly raw insecurity about how Tesla might be perceived if accurate data were revealed to the public. I happen to believe that Tesla is ready for the prime time and am confident that Tesla's sales numbers can stand to scrutiny, but apparently not many here share my outlook. I hope that Tesla's management will someday move forward on this as a gesture of confidence. I think if Elon Musk were to lead on this, shareholders would come along and feel pretty good about it. In the meantime, I don't think I'll press this issue any longer as little good comes from stiring up insecurities. I wish us all the best for quarterly results.

I don't believe it is a matter of confidence, at least not an issue of Tesla or Elon Musk's confidence in themselves or their products. Rather, the street is neurotic enough as it is and Mr. Musk has stated many times that the street is a distraction. Releasing this data only increases the distraction as the market overreacts in both directions. With a single product line being made in a single factory on a build to order basis will little to no inventory to buffer production changes, this will have outsized impact on Tesla. Brand perception is extremely sensitive, especially with these kinds of products. Many successful large companies do not release this data in this manner. Hardly anyone is seriously clamoring for Apple/Samsung to release monthly sales data on a per sales territory basis, nor do we have consistent order information either. Anyone that asks is politely but firmly shut down. Historically the auto industry is a bit different due to the dealership model where there is inventory sitting on dealership lots waiting for buyers. Tesla's business model is very different in that regard.

 

[AudubonB]

I have been following this thread closely, and I have not read a single argument to weigh against my suggestion of releasing sales - not "deliveries" and not "orders" and very, very definitely not geographically-parsed - daily.

 

[chickensevil]

I have been following this thread closely, and I have not read a single argument to weigh against my suggestion of releasing sales - not "deliveries" and not "orders" and very, very definitely not geographically-parsed - daily.

What then is your definition of a "sale" because according to Tesla THEIR definition of a sale is a delivery. At the point that you push the "take delivery" button your order is finalized and Tesla counts the value of your car toward their bottom line and everything clears up.

If they published daily "sales" (that is, by Tesla's definition a delivery), then you would still run into the same issue I outlined above (as others have) which is that you would have wild swings month to month. Because October production generally gets the majority sent over seas which takes an extra 4-6 weeks (so November/December overseas delivery timeframe). What you would see, even if it was their global number for daily deliveries, is that you would have 1 or 2 deliveries per day in October (ok, that might be a little too low, but lets just go with it), then jump to 10-20 per day in November, then jump to 100-200 per day in December (yes, these are all extreme fictional numbers that wouldn't likely happen like that, but is shows the point I am trying to make).

So you would still have the doom and gloom around October numbers, people would continue to drive that in Novemeber, and then finally in December you would get full quarter clarity which would confirm or very likely deny all the FUDsters slinging their crap. So all you gain with this is removing that guesswork from December 31st to whenever in Feb they release the Earnings numbers... But it does NOTHING for the Oct/Nov timeframe (and most of December).

So then you move into the new quarter, and while Tesla is talking about their ER for Q4, in February, you would have all these Analysts ignoring all that information and slamming the Conference Call with Questions about "why are January numbers so low" "is this a sign of the end of times for Tesla demand" "What are you doing to raise demand further"... hmmmm that sounds awfully familiar to what we hear right now... So what exactly has changed?

 

[AudubonB]

I've no problem with equating sales and deliveries - whichever metric accounts for the money in TMC's bank. It seems to me, however, that you are glossing over jhm's revelation that my proposed exercise completely deflates the bears' "gotcha" tool, for two reasons: first, because although the prior one, seven, thirty days' numbers were X rather than X+1....tomorrow's numbers will alter that!; and second, the constant release of data means that viewers stop focusing on those data. That latter is extremely powerful.

If you would like to look at it another way, try this: what is it in your analysis that logically​ precludes your suggesting to Tesla they report sales numbers only every - oh, say - ten years?

 

[chickensevil]

I've no problem with equating sales and deliveries - whichever metric accounts for the money in TMC's bank. It seems to me, however, that you are glossing over jhm's revelation that my proposed exercise completely deflates the bears' "gotcha" tool, for two reasons: first, because although the prior one, seven, thirty days' numbers were X rather than X+1....tomorrow's numbers will alter that!; and second, the constant release of data means that viewers stop focusing on those data. That latter is extremely powerful.

If you would like to look at it another way, try this: what is it in your analysis that logically​ precludes your suggesting to Tesla they report sales numbers only every - oh, say - ten years?

Because they wouldn't be able to get away with it as a public company? I'm pretty sure in one form or another you have to report your revenues in your Quarterly reports as a Publicly traded company. I do think there would be less stress on them to pump out NA sales and such at the end of the quarter if they weren't trying to make certain deadlines every quarter. So going to a yearly based number would likely help in that regard... but they can't because I am sure the SEC wouldn't allow it.

Technically Tesla doesn't have to tell people anything ever until the Earnings releases. So really we should all be thankful we get the information we do get inbetween earnings. Which to me says they are not giving out these numbers for a reason because clearly Elon likes to talk about many different things at times when he probably shouldn't. So if they feel it is necessary to hide that data then I trust that judgment. Which is why I said I am not opposed to them releasing the numbers whenever and however fast that they want to, but let them be the ones to decide that since they should *hopefully* know more than we do about such things.

 

[Krugerrand]

And the answer from the horse's mouth is: No, you can't have those numbers. And if you don't like it, take your money and invest it in a company more to your liking because Tesla isn't changing.

 

[Auzie]

I've no problem with equating sales and deliveries - whichever metric accounts for the money in TMC's bank. It seems to me, however, that you are glossing over jhm's revelation that my proposed exercise completely deflates the bears' "gotcha" tool, for two reasons: first, because although the prior one, seven, thirty days' numbers were X rather than X+1....tomorrow's numbers will alter that!; and second, the constant release of data means that viewers stop focusing on those data. That latter is extremely powerful.

If you would like to look at it another way, try this: what is it in your analysis that logically​ precludes your suggesting to Tesla they report sales numbers only every - oh, say - ten years?

My preference is for the opposite 'Tesla reporting world' of the one you propose.

I really like the 'suspense', I enjoy reading about people's speculations of mysterious irrelevant numbers, and I prefer to see bears inflate their 'gotcha' tool rather than deflating it. Bears being wrong works charm for TSLA. I never had so much fun with any other stock.:wink:

Tesla management will do whatever serves their business best. Elon does not tweet about NA sales every month, but he tweeted about NA record sales number in September when there was a purpose for such information to be released.

 

[chickensevil]

And the answer from the horse's mouth is: No, you can't have those numbers. And if you don't like it, take your money and invest it in a company more to your liking because Tesla isn't changing.

Specifically here is the piece from the transcript that covered exactly this question:

No, sorry. The last question was sort of a demand-related question as well and it really took pains to emphasize demand is not our issue. Production is our issue and being too perfectionist about future products those are legitimate things to be concerned about, but not demand. We have more demand than we can really address and there is a lot of things, levers we could pull to increase that demand which we're not pulling. So, it's really not an issue.

Part of the reason why we don't release the monthly deliveries number is just because it varies quite a lot by region and then the media tends to read all sorts of nonsense into deliveries. We'll have like 1000 cars reach a country one month and none the next month and then people -- or like 100 the next month trickle in or something because those were the numbers that were registered in one month versus the next and people say Tesla sales dropped by a factor of 10.

The boat arrived in January and not all the cars got registered in January and some got registered in February, then in March, it's back up again and so people read in all these things which are -- they assume deliveries are proxy for demand which is not the case. It is the case for other car companies, but in our case it really needs be parsed into orders and deliveries. Then bear in mind, there are lots of things we can do to amplify orders.

Orders is not a true measure of demand, it is just a measure of that's the amount of stuff we need to do to meet our production and delivery number. So if we released orders, people would try to read the tea leaves and say demand for Tesla is growing or dropping. No, we're just not pulling the levers that we could pull because there is no point in trying to amplify demand substantially beyond our ability to produce it and deliver it. That would just make people upset.

I am sorry if this displeases some of you here, but I really don't think sales is the real issue right now the only thing that is going to hamper sales is production. And when we are sitting here quipping over the number of deliveries being 7600 or 7800 or 8000... Management is up there thinking "oh man, how are we going to ramp production up to hit 50,000 a year!"

I think we all need to remember to keep a little perspective here and not get too focused on the day to day/month to month... because the real target is how they are progressing toward Model 3 which isn't for another few years.

 

[jhm]

Deliveries are operational metrics. It not a demand metric. Bears have tried to frame deliveries as a demand metric, but that is not what they are. As an operational metric, deliveries show us how well Tesla is serving their customers. It tells us how balanced and reliable their logistics are. It tells us if certain customers have to wait a long time for an infrequent ship to arrive. Why would you send 1000 cars to one country in one month and then wait another two or three months to send another shipment of a 1000? Would not it be a better customer experience if one shipment of 350 cars came every month? Demand is squandered when shipments are not timely or reliable. Our international customers deserve better. Let's stop allowing the bears frame this issue for us. The issue is how well Tesla is delivering for our customers?

 

[RobStark]

Why would you send 1000 cars to one country in one month and then wait another two or three months to send another shipment of a 1000? Would not it be a better customer experience if one shipment of 350 cars came every month?

Because it is cheaper to ship in bulk and pass on those savings to international customers.

 

[Krugerrand]

Because it is cheaper to ship in bulk and pass on those savings to international customers.

As well it saves Tesla money and time to batch for those ships once every 3 months than every month: as per their production issues.

 

[jhm]

Is it now time to start reporting month sales numbers?

Tesla is winning now.
Model 3 could become the top selling car in the US within a year.
Tesla is straining delivery network in many areas, so they want to avoid packing big numbers into the end of each quarter.
Authoritative sales numbers will undercut FUD that wants to undermine Tesla's success.

When I first opened this thread, I was amazed that so many Tesla fans thought publishing monthly numbers was a bad idea. Elon himself settled this. But when you are winning, why don't you want everyone to know what the score is? Perhaps it is time.

 

[strangecosmos]

Elon addressed this many years ago in Detroit when asked by a journalist. He said he didn’t see how it benefitted Tesla. But he joked that he did see how it benefitted journalists: they could write three times as many articles.

I would enjoy getting the daily or even hourly sales figures if I could, but I’m not sure it would actually be useful. What actionable difference does it make if I get daily, weekly, or monthly data instead of quarterly data? What action does the monthly data enable me to take that the quarterly data doesn’t?

 

[jhm]

Elon addressed this many years ago in Detroit when asked by a journalist. He said he didn’t see how it benefitted Tesla. But he joked that he did see how it benefitted journalists: they could write three times as many articles.

I would enjoy getting the daily or even hourly sales figures if I could, but I’m not sure it would actually be useful. What actionable difference does it make if I get daily, weekly, or monthly data instead of quarterly data? What action does the monthly data enable me to take that the quarterly data doesn’t?

The difference is that now we actually want journalists to write three times as many articles on how strong Tesla sales are. How do you feel about investors checking the likes of the Bloomberg Model 3 Tracker or a Twitter bear like Skabooshka on a daily basis? Investors do search out daily information, but without timely and authoritative reports from Tesla they go to less credible sources for daily info. So the action this enables you on a monthly basis is to put a reality check on less credible daily reporting, You can fight FUD with timely facts.

 

[jhm]

Yesterday, Musk retweeted an article claiming that Tesla models were the 1st, 2nd, and 3rd best selling EVs in the US for August. Bloomberg reporter Dana Hull asked if he would confirm the accuracy of the article he had retweeted. Elon failed to respond to this confirm ask.

Elsewhere on Twitter, the same day someone challenged Dana on why she would report on how Model 3 sales were topping charts in August. Her response was simply that Tesla only releases this information quarterly.

So what should Dana as a reporter do? Should she right a story about how other journalists are guessing at Tesla sales numbers, but which Tesla declines to confirm as accurate? Or should she not an article on Tesla sales until the quarterly numbers come out in early October?

Tesla and Space X is Dana's bear, so she is writing stories on nearly a daily basis. If she is not writing about August Tesla sales number right now, then she has to dig into other sources for her reporting. So while Tesla refuses to confirm what would make an excellent headline for both Tesla and Bloomberg, she has little choice but to turn to less credible source to dredge up something else to write on.

And we are amazed that so much FUD finds its way into the news feed today.

 

[strangecosmos]

If you are investing on, say, a 5-year time horizon, what does it matter whether the articles get written today or in three weeks when Tesla announces Q3 production and deliveries?

I think in the factory Tesla has a big screen that tracks the total weekly number of cars produced minute by minute. If Tesla said, hey, we’re going to put that counter up live on our website, I wouldn’t have a problem with that. I would find it cool and interesting, and I would probably check it a few times per week (against my better judgment).

But I also don’t have a problem with Tesla announcing production on a quarterly basis. The proper unit of time with which to evaluate a company isn’t months, or even quarters, but years. So announcing deliveries and production four times a year seems like plenty.

Almost everything that happens on Twitter, Facebook, forums, and the daily churn of news articles doesn’t matter. It’s just noise, like the waves crashing up against the shore. It will be there one moment and gone the next, leaving no trace. You can safely ignore it and everything will be fine.

What makes a lasting difference is slow, quiet, focused, deliberate thought and action over a long, long period of time. Ashlee Vance made a lasting impact on public consciousness by writing his book. He affected me on a deep, personal level, and I feel grateful to him for that. But I doubt whether his tweets or even most of his articles will have any lasting effect. Probably, they are inconsequential, like most of the things that get tweeted or published.

I’ve written 56 articles about Tesla and I’m not sure if any of them matter at all. They were fun for me to write, and important for me personally, but did they have any meaningful impact on the world? I don’t know, probably not. I generated a lot of ad impressions for Seeking Alpha, and gave a lot of recreational commenters stuff to argue about, but that might be it.

The annoying truth is doing anything that matters is really hard and takes a long time. Anything easy and quick, like tweeting, or dashing off a sassy essay, most likely doesn’t matter. The world doesn’t move because of people arguing on the Internet. As much as it feels like the stakes are high. The wave rolls back into the ocean whether you push it or not.

What did people do hundreds or thousands of years ago that still matters today? Or even decades ago? They seem to be mostly things that took quiet focus over years and decades. From this perspective, the difference between one month and three months looks so small. Why not take the time you would have spent tracking and chronicling the monthly blow by blow and plough that time into an effort that will accumulate over time? In 5 years from now, which will matter more?

Cal Newport on doing Deep Work and escaping social media - The Ezra Klein Show

 

[RobStark]

Once Tesla starts shipping Model 3 overseas there will be big variance in monthly US deliveries as well as European and Chinese deliveries. Massive FUD opportunities if Tesla reports monthly deliveries. "Tesla US sales down 50% this month, is Tesla demand Evaporating? Is the EV market Saturated ? "

IMO Tesla should not report monthly sales until they have a fully ramped up Chinese and European GF.

 

[adiggs]

Once Tesla starts shipping Model 3 overseas there will be big variance in monthly US deliveries as well as European and Chinese deliveries. Massive FUD opportunities if Tesla reports monthly deliveries. "Tesla US sales down 50% this month, is Tesla demand Evaporating? Is the EV market Saturated ? "

IMO Tesla should not report monthly sales until they have a fully ramped up Chinese and European GF.

An easy way to counter this dynamic is to start reporting monthly production, probably broken up (if the reporting started today) as the S/X total and the 3 total. Don't break it down by targeted geography. That keeps the reporting simple from Tesla's point of view, provides some facts for the reporters to write about on a monthly basis, and the tea leaf readers will still have something to do as they pour over registration information to figure out which geos got the cars.

Oh - and probably round or floor to the nearest 100, and don't report information for models until they have at least 100 units of monthly production (thinking ahead to Roadster - 1200/year is worth reporting, barely - less is better to leave as "early units are delivering, but production is still ramping and debugging").



In the future when another factory is online and producing in volume, the monthly production report could include production breakdown by geography of production (Americas production = <blah>; Europe production = <bleh>; Asia production = <blooh>). Of course some or a lot of those vehicles will be exported out of their production geography, but a lot them will stay put.


So I find myself persuaded by @jhm's observation - if you're winning, why not provide the regular authoritative information reporters can use for headlines and articles that will be good for those organizations, and will also be good for Tesla. Tesla can certainly use the good press and I think @jhm is correct in the fundamental assumption that if Tesla is reporting the data, then reporters will report on that data. That will mean good and bad articles, but mostly good as of today, and I know I believe that will tend to continue to be good as 3 continues to ramp and starts to deliver overseas.

And keep it simple and avoid all of the delivery timeline and geography delivery nuance and ties closer to financials people want - don't report any of that. Cars out the gate at Fremont (today) regardless of where they're going and when they'll get there is an easy number to report.

And it will also help emphasize and start building the expectation that Tesla is moving to a more even flow of work - balanced deliveries in all geographies, all quarter long, rather than ebbs and flows based on quarter end financial reporting (to which I say - hallelujah).

There will admittedly be a lag between production as reported in the monthly figures and the delivery, but it'll also start becoming increasingly clear that this month's production is next month's registrations (or registrations over the next 2 months), further emphasizing how rapidly Tesla converts a built unit into a registered / purchased vehicle (relative to the rest of the car industry). This latter story will take several months or quarters to be well understood, but the production data will provide the baseline to calculate this latter info, and will become another part of the Tesla business model story that can be reported.

 

[adiggs]

Adding on - I just went back and saw the thread title. I would NOT like to see Tesla trying to report monthly SALES numbers. Investors and finance types would like to see that, but that's a much more difficult number to figure out and report on.

And it will tend to be very geography centric, if for no other reason the overall number will be drawn from geographic level data.

I WOULD like to see monthly PRODUCTION numbers. That's an easy one - it's a single (or 2 or 3) value and today it can all be found at the Fremont factory. Monthly production numbers will also include cars built for crash testing, or store / inventory cars, and all the other purposes that don't directly tie to a customer delivery.


Production numbers get you the lion's share of the benefit, some degree of front running information (strong correlation with next month's sales / deliveries), and its easy.

In a demand constrained environment, as Tesla is in, production is really what we want to know anyway. The geography of delivery and the timing is important detail work, but that's also all that it is.

 

[jhm]

Hey, @adiggs, thanks for recognizing my points. I'm not too concerned about the specific content of a monthly report, what granularity, what metrics. But any reasonable set of metrics that can help narrow the degrees of freedom on uncertainty would help. Consistency is key. With consistent and timely reporting, Tesla will gain in credibility.

This is why it is important to keep reporting even when the numbers might not look so good. Musk makes a lot of claims but is not so consistent in reporting back with the actual results. If the results are super, he want to blurt it out on Twitter. If they are less than super, he'd rather avoid saying anything. Have we heard back on those safety reports? The lack of reporting actually undermines his credibility and invite critics to attack. On the other hand, if one can be relied upon to deliver reports whether super or not so super, it invites trust and credibility. If fact Musk tends to be a straight shooter type, so consistent reporting would be quite compatible. The effort to always spin things into the most favorable light is probably not worth it. But to say in a timely and consistent manner, "here are the numbers," would make much more credible any explanation that may be required to understand what the numbers mean.