Thanks for your work, Deagol. I think there will be trouble if you are correct in your estimate that AAPL will guide $0.93 on Tuesday.

[quote author=“lumi”]Thanks for your work, Deagol. I think there will be trouble if you are correct in your estimate that AAPL will guide $0.93 on Tuesday.

Compare with deferred revenue, eps guidance is not that relevant.

Thanks, Deagol.  Agree with “AAPL is once again a strong buy at these depressed levels (anything close to $160). Be careful on the short term choppiness.”  Only guys on margin and near-term options traders need to worry greatly.

[quote author=“deagol”]Strong Buy

[list]Revealed iPhone and Leopard sales appear inline with model’s estimates.
Tweaked model to account for Mac Pro and MacBook Air (higher AUPs), Time Capsule, Apple TV, and movie rentals.
Added new tab to compare historical (last two years) stock prices vs. model’s targets/fair value.
This update gets archived as the final F08Q1 version.
Guidance seems inline to slightly below Wall Street consensus estimates.
AAPL is once again a strong buy at these depressed levels (anything close to $160). Be careful on the short term choppiness.
Good luck to everyone next week. GO APPLE![/list:u]

3m ending on Dec-2007:

Deagol, thanks for your latest estimates.

I have a question… Would you explain to me about US web order daily rates and
Monthly interpolation/proj….?  How do you come up with these? Thanks.

Also, what do you think of Gartner & IDC repot of US computer shipment?? Those numbers really make me nervous. They, of course, could be off but how much?

Apple gd: $9.2 b rev, $1.42 EPS
Analysts: $9.44b rev, $1.61 EPS
My estmt: $9.71b rev, $1.80 EPS
25M iPods, 2.25M Macs

3m ending on Mar-2008:
My estmt: $7.21b rev, $1.18 EPS
Analysts: $6.97b rev, $1.09 EPS
Apple gd (my est): $6.85b rev, $0.93 EPS

12m ending on Sep-2008:
My estmt: $33.20b rev, $5.55 EPS
Analysts: $31.77b rev, $5.13 EPS

Valuation (25 Ч fwd EPS):
Mar-2008 (fair value): $160, + 0% upside (Mar-2009 EPS $6.42)
Dec-2008 (12m target): $200, +25% upside (Dec-2009 EPS $8.01)
Analysts mean target : $214, +34% upside

http://www.esnips.com/web/aaplweborders

Sorry my post to Deagol messed up.
Will post again here,

Deagol, thanks for your latest estimates.

I have a question… Would you explain to me about US web order daily rates and
Monthly interpolation/proj….? How do you come up with these? Thanks.

Also, what do you think of Gartner & IDC repot of US computer shipment?? Those numbers really make me nervous. They, of course, could be off but how much?

[quote author=“FUJI”]Sorry my post to Deagol messed up.
Will post again here,

Deagol, thanks for your latest estimates.

I have a question… Would you explain to me about US web order daily rates and
Monthly interpolation/proj….? How do you come up with these? Thanks.

Also, what do you think of Gartner & IDC repot of US computer shipment?? Those numbers really make me nervous. They, of course, could be off but how much?

Sure I can, no problem!

Every time a customer places an order on the Online Store, Apple generates a unique and sequential order number. A few years back, a poster from the yahoo board (back when that board was relatively usaful) realized that these order numbers were sequential and with no gaps in between orders. So, this amazingly lucky finding (given Apple’s penchant for secrecy) allowed us to calculate the total number of orders placed online, in real time, for any quarter, by simply placing an order at the end of each quarter. As you can imagine, the number of online orders has a pretty high correlation with total revenue, so given some historical data it’s quite simple to derive a revenue estimate the same day the quarter closes (or even a day close to it and then interpolating for the precise date) instead of having to wait 3 weeks for the earnings report.

So, the “US web order daily rates” are calculated for any given period by dividing the number of orders over the number of days in between them, giving a real, factual figure of how many orders, on average, are being placed per day for any given period. The “monthly interpolation/projections” are calculated simply by interpolating in between two samples for the order number that would have been placed in the end of every month in order to calculate the rates for each month of the year and from that derive a seasonality pattern, which is then used to project (through a regression of the historical order rate) into the future, adjusting this regression by the corresponding deviation from the regression line given by the month in question. This then provides a method of tracking the real time sampling of real orders against this projection derived from the historical growth trend and seasonality, and from that comparison evaluate whether sales are following the trend, falling behind, or accelerating.

Hope this explanation helped, and feel free to ask me any questions you may still have.

As for Gartner and IDC, I suspect they have changed their methodology or something, as those figures are not making any sense to me.

PS: Here’s a more extensive explanation of the web orders model from way back when I got here.  

Deagol, thanks for your explanation.
the formula you came up with is pretty amazing. Though I don’t fully understand each formula, I got the overall picture.
THANKS!!!

Re: Deagol’s Q1 2007 Spreadsheet + Estimates

[quote author=“deagol”]Hi enature, nice to talk to you after all this time. Good to know you’re still bullish. We’re almost there for your target of “$300 in a couple of years.”

I’ll try to add $1 billion in revenue from your suggestions, which is what is required to get to an EPS of 2 bucks.

[quote author=“enature”] Deferred revenue from iPhone in this quarter will count in Q1

You mean approximately 1/8 of the deferred revenue, right? sorry if I’m missing your point. I have 333 million for iPhone revenue, but even if I double the sales to 5 million units, the iPhone revenue for this quarter doesn’t go over 500 million. I can’t sensibly justify adding more than $150 million revenue from iPhone sales upside.
[quote author=“enature”]... and iPod margins and revenue per iPod are higher in Q1.

According to the CC, more iPods in the product mix are one of the reasons for their guiding margins slightly lower. I’ll consider a higher unit price… let’s see, last quarter was $159/unit, and last year’s Q1 was $163/unit. I don’t see AUP’s going up y/y, so $163 is 3 bucks higher than my current estimate, times 26 million units adds less than $80 million to revenue.

So, here’s as high as I could get it (million dollars):
+250 (Mac)
+150 (iPhone)
+ 80 (iPod)
——
+480 ...we didn’t get half way there. Do you have any other rocks to look under (preferably bigger ones )?
[quote author=“enature”]
I see EPS closer to $2, which already puts the stock above $200 under old P/E. If you count P/E expansion then high $200s are justified.

I’d like to see the details of how you go from the $2 EPS to the $200 target under old P/E. Not questioning your reasoning here, just curious what specific figures you’re using. In my case, the $181 target is 25 times F09 EPS estimate of $7.25.

Thanks!

deagol,

please, excuse my tardy response, but I have more info now to reply.
As Tommo said iPod touch increases ASP. iPhone is a very high margin product. 4 mil iPhones + subscription fees fatten margins.
The ratio of the total revenue to the US web orders is higher because 1) more retail stores, more 3rd party distribution points 2) more international sales.

Why do I say that $2EPS firmly supports $200 price? Because my simply rule of thumb has been working for AAPL very well in the past. It goes like this:

Multiply ESP for any quarter by 100. That’s a fair share price for 25 times earnings. Why? Because there are 4 quarters in a year, 100/4=25.

One might say: “But earnings are seasonal, while you extrapolate current earnings on the entire year.”
Well, that’s the beauty of this rule of thumb - it takes into account seasonality. You can apply to actual ESP or expected EPS, but it worked well for actual ESP.
Well, 25 multiple is too conservative, so I actually was adding up to $12 to come up with fair share price.

So $2EPS means $200-$212 share price as a minimum. Why minimum?
Because something paradoxical has been happening to AAPL.
Quarter after quarter its actual earnings grow 50-80% range, but the street every time disregards it as an aberration and assumes 25% growth. This comfortably falls in line with AAPL uPODing. Apples undeguidance keeps P/E multiple as 25% growth.
This striking dichotomy between the actual EPS growth rate and what is assumed by the street cannot last forever. At some point, the street will open up to real 50-80% growth. This will ballon P/E to the point where my $300 price target will look dirt cheap.
Tommo on many occasions stressed this point. It might happen this Jan or some time in 2008, but be ready for a parabolic explosion in share price.

Here is my EPS estimate based on web orders pulled from deagol’s data.

I round web orders to two digits for simplicity, followed by EPS

for Jan 08 earnings, 34, EPS ?
for Oct 07 earnings, 22, EPS 1.01

for Jan ‘07 earnings, 28, EPS 1.14
for Oct’06 earnings, 15, EPS .62

If I simply extrapolate using web orders from last Oct, then the expected EPS is

1.01*(34/22)=1.56.

But simply extrapolating is wrong. Last year we had a boost from Oct to Jan. Precisely, the boost is following. Jan ‘07 EPS “should have been” 28/15*.62=90.

However, the actual EPS was 1.14. Thus, the boost is 1.14/.90=26%
(anyone can recall if 1.14 included one time adjustment?)

So we assume the same boost, we get 1.56+1.56*26%=$1.96

So EPS of $1.96 is expected. However, say I assume 1/2 of the boost, 13% to be conservative.
EPS is $1.76. But this does not include EPS from all that deferred revenue.
Anyone calculated how many cents iPhone from last quarter adds to this Q?
If we add that to $1.76, we get the rock bottom EPS estimate

[quote author=“enature”]
Anyone calculated how many cents iPhone from last quarter adds to this Q?
If we add that to $1.76, we get the rock bottom EPS estimate

Please see the Apple’s Magic Money topic .

[quote author=“enature”]
If I simply extrapolate using web orders from last Oct, then the expected EPS is

1.01*(34/22)=1.56.

This is exactly what Caligula does. IMO it’s too simplistic as it doesn’t account for higher revenue outside of the US online store (retail, international); margin expansion on higher revenues; the effects of any product transition/introduction; the effects of the Christmas season product mix; iPhone revenue recognition and revenue sharing; higher interest and other income; etc.
[quote author=“enature”]
(anyone can recall if 1.14 included one time adjustment?)

It didn’t, but the fourth quarter EPS have included “one time” tax benefits over the last three years, as follows:

I’m not sure how this affects the projections you used.

[quote author=“deagol”].........
So, the “US web order daily rates” are calculated for any given period by dividing the number of orders over the number of days in between them, giving a real, factual figure of how many orders, on average, are being placed per day for any given period. The “monthly interpolation/projections” are calculated simply by interpolating in between two samples for the order number that would have been placed in the end of every month in order to calculate the rates for each month of the year and from that derive a seasonality pattern, which is then used to project (through a regression of the historical order rate) into the future, adjusting this regression by the corresponding deviation from the regression line given by the month in question. This then provides a method of tracking the real time sampling of real orders against this projection derived from the historical growth trend and seasonality, and from that comparison evaluate whether sales are following the trend, falling behind, or accelerating.

Hope this explanation helped, and feel free to ask me any questions you may still have.
.......

deagol, I am taking the liberty of piggybacking on your gracious offer to Fuji.

I am imagining you cannot track actual average-order-size.  But can/do you track a substitute for that?  For example “total revenue/number of online orders”.  If so is the rate of change fairly constant or is it increasing?

Thanks.

[quote author=“capablanca”]For example “total revenue/number of online orders”.  If so is the rate of change fairly constant or is it increasing?

You’re correct, that is all I can do. And yes, that ratio’s been growing (about 10% annual growth over the last 2-3 years). I expect it to accelerate to 15% over the next 2 years while the iPhone subscriber base grows and the revenue recognition builds up (because today’s iPhone orders represent future revenues up to some point when the ratio stabilizes).

Nice work deagol.  Your EPS and Revenue numbers were quite on target.  Must give credit where credit is due!

Well my friend you were off by only .02 not bad.

I was hoping you would be way off, but we missed this one big time.

Congratulations.  We will have to pay more attention to you.

[quote author=“omacvi”]Well my friend you were off by only .02 not bad.

I was hoping you would be way off, but we missed this one big time.

Congratulations.  We will have to pay more attention to you.

i’m sorry.