Quick thoughts on Microsoft and Amazon in the cloud

At the end of last year, I wrote a piece for Techpinions Insiders about what to expect for Amazon in 2015, as part of a series on major tech companies. As part of that piece, I wrote the following about Amazon’s AWS business:

AWS has been one of Amazon’s big success stories over the last several years, generating higher margins than e-commerce and growing extremely fast. But growth faltered a little in 2014, as competition, from Google and Microsoft in particular, intensified. The basic storage and infrastructure services are becoming rapidly commoditized, with plummeting prices and little real differentiation. As such, both margins and differentiation will move to what these companies build on the basic services; hence Amazon’s launch of Zocalo and other enterprise tools that sit on top of AWS. But here it is going up against Microsoft’s traditional stronghold and Google’s increasingly capable offerings. End user software hasn’t been Amazon’s strong suit, but it’s Microsoft and Google’s bread and butter. I remain skeptical of Amazon’s ability to successfully compete in this area. Meanwhile, others not in the cloud storage business are also building their own competing platforms at this higher layer, including Box, Salesforce and others. If AWS is to become the highly profitable core Amazon has always lacked, it needs to successfully compete at this layer as well as the basic services it has provided historically. It’s not clear to me Amazon will be any more successful at this in 2015 than it was in 2014.

This week, Microsoft reported its earnings for the December quarter, and among other things Satya Nadella said this about Microsoft’s own cloud business in response to a question about cloud margins:

Overall, the shift to the higher layer services is the real driver here, which is obviously Office 365 and its various levels is one factor. The other one is what I talked about in the Enterprise Mobility Suite, that’s really got fantastic momentum in the marketplace because the solution has really come together and is fairly unique, as well as Dynamic CRM.

So these are all got a different profile in terms of margin and they are all now pretty high growth businesses for us. So when you think about our cloud, you got to think about the low-level infrastructure. Even there we now have premium offerings and then we have higher level services. So that aggregate portfolio is what helps us move up the margin curve.

I juxtapose these two quotes – one about Amazon and one about Microsoft – because I think today’s announcement by Amazon of new enterprise email services plays directly into the challenge I described, and which Microsoft seems to be managing much better. It continues to be critical for both companies (and Google) to migrate their way up the cloud stack to the higher-layer services (as both I and Nadella called them), but Microsoft is already there, while Amazon continues to try to compete in a space I’m really not sure they can. We’ll see what AWS results look like tomorrow, but I expect this to be something that comes up on the call.

Thoughts on Apple’s Q4 2014 earnings

Notes: this is part of a series on major tech companies’ Q4 2014 earnings. All past earnings posts can be seen here, and all earnings posts for Q4 2014 can be seen here. For the sake of easy comparisons and transparency, I always use calendar quarters in my analysis. Hence, Q4 2014 in this and every post on this blog means the quarter ending December 2014, even though some companies (Apple included) have fiscal years that end at other times of the year. 

A blowout quarter

The hardest posts to write are often the ones where it’s utterly uncontroversial that the results were astonishingly good, and that was definitely the case with Apple’s record-breaking earnings today. So instead of hashing over the same stuff as everyone else is, I’m going to try to pull out a few possibly overlooked data points. Apple changed its reporting structure in a couple of ways this past quarter, and that gave us one or two new insights while also sadly burying some data points and obscuring others. I’m going to be working through the revised numbers over the next 24 hours or so, and will be issuing my quarterly deck for subscribers once the 10-Q report is out, as that’ll fill in some gaps in the current data. I may well do another post on the earnings at that point too.


I tweeted about the iPhone ASPs as follows shortly after the numbers came out:

In some ways, that about sums it up. But of course that chart shows two sets of ASPs, going in dramatically different directions (as I indicated they would in my Techpinions earnings preview post on Monday):

  • iPhone ASPs rose in Q3, but even more dramatically in Q4, largely thanks to two things: the iPhone 6 Plus, which raised the base price of the top-end iPhone model by $100, and the introduction of a 128GB model, which raised the top-end price as well. The combination of these two conspired to lift ASPs $50 above last year’s number.
  • iPad ASPs continue to fall, on a fairly predictable slope, over the last few quarters, enabled conversely by a lowering of the entry price for iPads.

The two charts below show the pricing moves behind those ASP trends:

iPad retail prices iPhone retail pricesAs you can see, the lowest price for iPhones has remained very stable for four years, while the highest possible price has risen $200 since 2010. But the iPad’s lowest selling price has fallen from $500 to $250 during that same period, while the highest price has barely changed. Given the lack of subsidies on the iPad, lower full retail prices translate directly to what consumers actually pay, whereas higher prices on the iPhone side are masked by carrier subsidies and/or installment plans in many cases. All this helps to explain why the iPhone ASP keeps rising while the iPad ASP keeps falling.

Retail’s varying importance in different geographies

Apple giveth and Apple taketh away when it comes to financial reporting. This quarter, Apple took away all visibility over the current quarter’s retail finances, as it rolled retail reporting into regional reporting. However, in so doing, it provided a wonderful insight into past retail performance on a region-by-region basis, something we’ve never had before. I’m curious to see whether it provides any retail-related financials in its 10-Q, but for now we’ll have to make do with this interesting data set. I’ve taken that historical data and generated the following chart, which shows retail revenues as a percentage of total revenues on a regional basis:

Retail as percent of revenuesThere’s obviously a huge variability by region, and this reflects a factor I’ve documented in the past. Here’s the old-style revenue split by region vs. the number of retail stores by region, which highlights the regions which have fewer retail stores than their revenue contribution would suggest:

Retail stores vs revenueAs you can see, the regions with the smallest percentage of revenue from retail are the same as those with the smallest number of retail stores relative to their overall revenue contribution, so this isn’t a big surprise. Japan comes bottom on both metrics. The next question is which of these two factors is the cause and which is the effect, given that there’s clearly a correlation. I suspect there’s some of each, but it’s also clear why Apple is investing so heavily in retail stores in China. It’s also clear why Apple is adding so many more retail stores outside the US than inside it (though that trend reversed a little in the last two quarters).

Growth remarkably diversified by region

One last data point: Apple’s growth this past quarter was amazingly widespread by region. Over the last five quarters Apple’s gone from pretty low overall growth back to roaring growth on a year over year basis. In some of those past quarters, one or two regions carried much of the overall growth, whether Japan for a couple of quarters last year, or China during almost all quarters. But this quarter was notable for just how broad-based that growth was by region, with every region but Japan making a pretty meaningful contribution to overall growth (and Japan suffering from tough comparisons with a very strong quarter a year ago rather than any poor underlying performance):

Year on year rev growth by regionApple provided some numbers around this on the earnings call, citing 22% revenue growth in developed countries, 58% revenue growth in emerging markets as a whole, but 70% revenue growth in China year on year. Unit shipments grew by an astonishing amount in the BRIC countries as a while too – I’m not 100% sure of the number but I believe it was 90%.

Divergent fortunes for Apple and other major phone makers

As I said, I hope to have more top-line analysis later on, but for now I’ll end with this thought: the contrast between Apple’s and most other phone makers’ numbers couldn’t be starker, perhaps most dramatically as it relates to Samsung: record-high shipments at record-high prices, generating record-high profits, just as other vendors are seeing ASPs plunge, shipments stall and and margins squeezed. There’s been so much skepticism for so many years about Apple’s ability to continue to make its unique business model work over the long term, and Apple continues to prove them wrong. I believe with the launch of the Apple Watch in April, HomeKit devices finally starting to ship in significant numbers in the coming months, CarPlay, Apple Pay and who knows what else that might arrive in 2015, Apple is simply reinforcing what’s becoming an incredibly strong, sticky, and growing ecosystem.

My thesis on Microsoft

It’s earnings season and I generally post a particular kind of post when that’s the case, including the Microsoft earnings post I did last night. Typically, I highlight a few key data points and analyze those, without stepping back to do a big-picture view on a company. But sometimes I worry that this leaves readers without a good sense of how I see the company in question and its prospects. So I wanted to do a follow-up post to yesterday’s in which I take a broader view and share my overall thoughts on Microsoft and its prospects. I’ll provide a list of links to previous pieces on Microsoft at the end of this post in case you’re interested in exploring any of this in more detail.

Windows PCs are in a long-term decline

A big part of how you see the future of Microsoft depends on what you think the underlying trend is and will be in Windows PCs, so let’s start there. I see two theories in the market at the moment, and which of these theories you subscribe to very much provides the lens through which you see Microsoft’s results each quarter. One theory is that Windows PCs are on either a stable or growing trajectory over time, and that any quarters in which there is negative growth are the exceptions to that rule. The other theory is that Windows PCs are on a downward trajectory over time, and that positive growth quarters are the exceptions rather than the rule.

For the following reasons, I subscribe to the second theory:

  • PCs as a form factor are now one of several that can be used for the purposes that once required a PC, with tablets and smartphones providing adequate computing power and capability for what many people need
  • PC hardware has reached the point where even those who see a need for a PC don’t perhaps see the need to upgrade it as frequently, because a PC from several years ago is still perfectly adequate, especially if relegated only to those tasks other devices can’t perform effectively
  • Competition is gaining ground, with Chromebooks taking significant share in education and Macs gaining share in the broader PC market, especially among college students and other key groups. As such Windows PCs will be an ever smaller share of total PCs
  • People are choosing platforms other than Windows in new device categories such as smartphones, tablets, wearables, smart TV devices, and so on. As these other platforms increase the degree of integration within their ecosystems, it will be harder for Microsoft to sell Windows PCs that don’t integrate as effectively with Android tablets or iPhones.

For all these reasons, I see a downward trajectory over time in sales of Windows in total, even accounting for the many different form factors Windows runs on. As such, last quarter’s poor performance in Windows sales is much more indicative of the longer-term trend than short-term headwinds. I see Windows 10 slowing the decline a little, but I actually think the free upgrades could stall or postpone new device purchases for some users, which may be counterproductive in the short term. I don’t see Windows 10 solving any of the fundamental challenges I just outlined. Continue reading

Thoughts on Microsoft’s Q4 2014 earnings

Notes: this is part of a series on major tech companies’ Q4 2014 earnings. All past earnings posts can be seen here, and all earnings posts for Q4 2014 can be seen here. For the sake of easy comparisons and transparency, I always use calendar quarters in my analysis. Hence, Q4 2014 in this and every post on this blog means the quarter ending December 2014, even though some companies (Microsoft included) have fiscal years that end at other times of the year. 

As usual, I’m going to run through a handful of charts and provide some analysis in this post. Subscribers to my quarterly company decks service will have received a slide deck with a much larger set of charts – this subscription is $10 per month for individual subscribers, and corporate subscriptions are also available.

The hardware business – some progress, some slippage

One of the themes last quarter was how the hardware business at Microsoft had progressed, and that’s a theme again this quarter. Surface revenues were the highest they’ve ever been in a quarter, crossing the $1 billion mark for the first time, and in addition the gross margins were not only positive for the second time, but significantly so if my calculations are correct:

Surface financials Q4 2014I suspect that, on an operating basis, the Surface line still loses money because of all the marketing spend involved, but I would guess it’s not a million miles away from producing a positive contribution margin at this point, which is enormous progress from the early quarters. I suspect, with almost all the revenue this quarter coming from Surface Pro 3 sales, that ASPs were around $1,000, and that this represents around a million unit shipments, with some accessory spending rounding out the rest. Chances are that’s not much higher than previous quarters.

Lumia sales also hit a milestone this quarter, crossing 10 million for the first time:

Lumia shipments Microsoft also reports non-Lumia phone sales, which dipped below 40 million for the first time in almost 10 years (including the former Nokia devices business). That business is now dropping fast. It’s hard to tell because Microsoft doesn’t provide ASPs for these two product lines, but based on my calculations, I suspect Lumia ASPs dropped from about $140 to closer to $120, reflecting the company’s focus on the low-end market. I continue to question the rationale for pursuing this end of the market, however. Microsoft’s thinking is that if it can capture these users, it can slowly migrate them up to more expensive devices along with Microsoft services, but I suspect that only a very small proportion of the users it’s capturing here will do so. They’re largely attracted by the low price and are also overwhelmingly in low-income demographics, making future revenue prospects poor.

Insight into Windows Phone financials

The “Windows Phone” reporting line has always been a fascinating one at Microsoft, and one I’ve dug into quite a bit in the past in search of nuggets of data. Two new nuggets came this time around, highlighting a couple of interesting points. Here’s my best estimate of the Windows Phone revenue line for the past two years or so, based on various statements in Microsoft’s SEC filings:

Windows Phone revenue Q4 2014What’s interesting here is the extent to which this revenue dropped when Microsoft folded Nokia’s devices unit into the company. Q4 last year saw over $1 billion in revenue reported in this line, but this quarter it was under $400 million. The vast majority of that drop – around 60% – was said to be related to the Nokia acquisition. That, in turn, suggests that Android licensing revenue was less than half of total revenues in this reporting line, contrary to many people’s belief that Microsoft derived more revenue from Android licensing than Windows Phone licensing.

The other interesting tidbit was that Microsoft reported a fall in cost of revenue for the Devices & Consumer Licensing business related to the Nokia acquisition as follows:

D&C Licensing cost of revenue decreased, mainly due to a $224 million decline in traffic acquisition costs, primarily driven by prior year costs associated with our joint strategic initiatives with Nokia.

The only thing I can think of here is that this relates to some combination of Windows Phone and online advertising, since TAC is a metric usually associated with online advertising businesses, and D&C Licensing houses those Windows Phone revenues we just looked at. It seems as though Microsoft may have paid Nokia a fee for the traffic it sent to Bing and other Microsoft online properties, and this was somehow reported in D&C Licensing as a cost related to the Windows Phone revenues. The scale of that fee is pretty significant – $224 million in a single quarter, or almost $1 billion per year, which should improve the margins for the online advertising business significantly.

Online advertising continues to diverge

Speaking of online advertising, the trends there continue to diverge, as they do at other major online advertising companies, between search and display revenues. I’m presenting below the percentage of online advertising revenues that comes from each of these sources, according to my calculations:

Online advertising revenue splitAs you can see, search advertising is now well over 75%, while display advertising is rapidly approaching 15%. I’ve talked before about these businesses, and my conviction that at this point Microsoft should simply pull out of the display advertising business, sacrificing a small amount of revenue but in the process bolstering its claim to be more sensitive to users’ data privacy concerns than Google, and this just reinforces the case. Bing is, at any rate, far more strategically important to Microsoft, especially given the integration with Cortana that’s critical to Windows Phone today and Windows 10 going forward.

Office Consumer transition continues

Another data point I’ve picked up on in the past is the performance of Consumer Office, which comes in two parts: the legacy Office model, and Office 365 Home and Personal. The two continue to move in opposite directions, but with the overall impact of falling revenues from this business:

Consumer Office Q4 2014The red line shows total revenues from these two, which continues to fall year on year as Microsoft goes through this transition. As Microsoft gives away more and more of the core functionality of Office for free in the consumer market, however, this trend will accelerate. One bright point in this area, however, was the bump in Office 365 users this past quarter. I’m not yet sure what to attribute that too, but it’s likely that a combination of Office on the iPad and the new, slightly cheaper, Personal option (launched in Q3 but available for the whole of Q4) both had an impact:

Office 365 subsAs the previous chart shows, the revenue stream associated with these subscribers is still small, but this is the future of Office in the Consumer market, so it’ll be well worth watching what happens to both the subscriber numbers and the revenue associated with them as Microsoft’s changing business model for Office kicks in (especially with the launch of Windows 10 later this year).

Lots more slides in the deck

Again, this is just a sampling of the data I collect and analyze on Microsoft on an ongoing basis. The deck that’s part of the subscription service has the full set, and is available now to subscribers. Sign up here if you’re interested (Paypal and credit card payment options available). A screenshot of the slides in the deck is below:

Screenshot 2015-01-26 20.21.22

Netflix Q4 2014 earnings

I previewed Netflix’s earnings along with three other companies’ earnings in my Techpinions Insiders post earlier this week. I’ll be doing the same for quite a few more companies this coming Monday. If you haven’t signed up yet, you might want to do so. It’s $10 per month or $100 per year to become a Techpinions Insider, and it includes weekly columns from me as well as several other analysts.

Today, I’m kicking off my formal coverage of Q4 2014 earnings with a post on Netflix’s results, which were published this afternoon. I’ve also created a deck with a more extensive set of charts on Netflix’s performance, which is available through this subscription, also for $10 per month. And with that ends the sales pitch. :)

I’m generally pretty bullish on Netflix – I think they’ve created a fantastic value proposition that’s clearly become the gold standard for online video services in the US, and they’re now rapidly expanding that model to the rest of the world too. And that’s a good thing, because growth in the US is starting to slow down. Interestingly, the official reason for that slowdown has changed since last quarter, when the company blamed it on the price increase it instituted in May 2014 (this quote comes from this quarter’s shareholder letter):

In October, we judged the leading factor of the similar decline in Q3 y/y net adds to be our May price change. Since then, with additional research, we now think that the decline in y/y net adds would have largely taken place independent of the price change. We’ve found our growth in net adds is strongest in the lower income areas of the US, which would not be the case if there was material price sensitivity. Additionally, we implemented a similar price change in Mexico during Q4, and saw no detectable change in net additions. We think, instead, the reduction in y/y net additions is a natural progression in our large US market as we grow.

Translation: we’re reaching the point in our US growth where there just aren’t as many new subscribers to sign up as there were in the past, and so growth is going to slow down. You can see this effect in the US streaming subscriber growth numbers:

Netflix year on year subs growthAs you can see, US streaming subscriber growth slowed in Q3, and slowed yet again in Q4, and is likely to continue to slow down in subsequent quarters. Whereas the company has been able to count on 6 million new subs per year in the US, it can no longer do so. All of which makes that international growth all the more important, and that’s accelerating rapidly. This quarter was the first time year on year growth internationally eclipsed domestic growth year on year, and that gap will continue to widen.

This expansion, of course, continues to happen at the cost of profitability, since entering new markets is very costly. As I’ve remarked before, Netflix seems to pick up the pace of international expansion every time the overseas business threatens to become profitable, and margins are currently on a downward slope in the international business:

Netflix margins by segmentHowever, that overall line masks two very different businesses: Netflix’s original overseas markets (Canada, Latin America, the UK, Ireland, the Nordic countries and the Netherlands) are now profitable on a contribution basis, but it’s entered so many new markets that those are dragging down overall performance. It’s now in 50 markets overall, and things are going so well that it’s actually planning to accelerate the rate of expansion such that it’ll be largely done by 2017, while staying profitable overall.

Of course, one major cost of both international expansion and continued growth in a saturating US market is marketing. Marketing costs have never been enormous at Netflix in comparison with content costs, but they are a substantial minority of total expenses. And they’ve been rising over time, both domestically and internationally. There are a couple of different ways to measure this and I’ve included both in the full deck of charts, but here’s one measure that I think is meaningful:

Netflix marketing costs per net addAs you can see from that chart, on a trailing four quarter basis, the cost in marketing terms to add a net new paid subscriber has been going up, both in the US and internationally. That cost is still under 10% of revenues in the US, but is over 20% in the rest of the world, reflecting the need to promote awareness of Netflix in many new markets. The good thing, though, is that the cost of revenues (largely driven by content acquisition) continues to shrink as a percentage of revenues, and cost of revenue per paid sub is now lower than revenue per paid sub even internationally:

Netflix international rev and CoR per subThat means that, as Netflix is able to dial back marketing and other non-content-related expenses, and as scale effects kick in, profitability should be eminently achievable. It’s also on track to become the first truly global video business – something that Apple has arguably come closest to so far, but where Netflix will eclipse its reach in the next year or two.

One last thing I wanted to touch on is a rather unique aspect of Netflix and its financials, which is the very predictable and steady progress in US margins. I don’t know of any other company that could make a statement like this (again, from the shareholder letter):

This year we plan to increase US contribution margins from 30% in Q1 to about 32% in Q1 2016 to about 34% in Q1 2017, etc. We’ll re-evaluate the margin progression model again in early 2020 when we hopefully achieve 40% contribution margins.

What other company talks about a specific margin target for five years out? And seems so likely to be able to achieve that given its past performance? Here’s that US streaming margin line again:

Netflix US streaming margin

Again, these and quite a few more charts are in the Netflix deck I’ve just put together for subscribers. Sign up here and you’ll receive it by email within a few hours, and others in the coming weeks and months as they become available. A screenshot of the slides in the deck is below:

Screenshot 2015-01-20 16.57.54

Why Windows 10 can’t fix Windows Phone

Ahead of Microsoft’s next reveal of Windows 10 later this week, lots of blogs and news outlets are talking up the promise of the new operating system to unify the PC and mobile versions and in the process “solve the app gap”. Most of what I’ve read, though, seems to look straight past a huge flaw in this whole concept, one that I’ve talked about several times in other places (notably in my in-depth Windows Phone report from a few weeks back – available here for free).  As such, I wanted to just quickly lay it out here for simplicity and clarity.

First, the theory: in Windows 10, Microsoft is creating a single operating system which will run across different form factors, with much of the underlying code shared and the rest tweaked by device type and size. This will allow developers to create apps which run 90% of the same code, with just some customizations for different device types and sizes. This, in turn, will allow Microsoft to tap into the vast number of Windows PC developers, who will now be able to port their apps to Windows Phone will very little additional work, which will drive a large number of new apps to the mobile platform, reducing the app gap relative to iOS and Android.

However, there’s a fundamental flaw in this argument, which is that the apps Windows Phone is missing simply don’t exist as desktop apps on Windows. Just think about it for a moment, and you’ll realize it’s empirically obvious: almost all the apps which are most popular on mobile are in one of these categories:

  • Games, which dominate the app stores, and tend to be mobile-only in many cases
  • Properties which exist as websites on the desktop and only exist as apps on the mobile side
  • Properties which are mobile-first and/or mobile-only, such as Instagram, Vine, Viber and so on.

But we don’t need to rely on gut feel here – it’s very easy to do the analysis. I’ve pasted below two small thumbnails which you can click on to expand to full size. They show tables for the top free iOS and Android apps as of today, according to App Annie. Against each of the apps I’ve completed several more columns to reflect the following data:

  • Is the app already in the Windows Phone store?
  • Is there a desktop app on Windows (any version, not just Windows 8)?
  • Is this an app which is actually a website rather than an app on the desktop?

I’ve then done some filters in the following columns to answer each of the following questions:

  • Of those apps which are not on Windows Phone today, are these available for Windows PCs today?
  • Of those apps which are not on Windows Phone today, are these available as a website on desktop?
  • Of those apps for which there is a desktop app on Windows today, are these also available on Windows Phone?

You can go ahead and have a look at the tables to see the results for yourself (they should open in a new window or tab by default):

Screenshot 2015-01-19 09.54.42Screenshot 2015-01-19 09.54.57

But here’s the summary:

  • Among the top 50 free iOS and Android apps, there is not one which is not on Windows Phone but exists as a desktop app on Windows
  • Among the top 50 free iOS and Android apps, there are a handful which exist as websites but not as desktop apps (almost all owned by either Google or Apple)
  • All of the top 50 free iOS and Android apps for which there is a Windows desktop app already exist as Windows Phone apps today.

In other words, if the theory is that sharing a code base across desktop and mobile will lead to desktop apps being ported to the mobile environment in greater numbers, within this sample at least this has no applicability at all. All the apps available on Windows PCs are already available on Windows Phone. A handful of the rest exist as websites on the desktop, but the vast majority simply don’t exist today on any flavor of Windows.

There are two important caveats here. Firstly, this analysis only looks at the top 50 apps, and a different pattern could theoretically emerge if one were to examine a longer list of apps. However, from what I’ve seen the patterns are broadly similar, and the same conclusions would apply. Secondly, this analysis focuses on the most popular apps, which are naturally dominated by consumer-facing applications and not those used in the enterprise. I do believe that there are cases where desktop apps exist for enterprises but not yet for Windows Phone, and in this case the theory behind Windows 10 may well have at least some applicability. But that’s a far cry from saying that Windows 10 will help to solve the app gap, which is fundamentally a consumer problem, not an enterprise one.

Having said all this, I’m very curious to see what Microsoft has to say this week with regard to the mobile flavor of Windows 10 in particular. I think it’s getting a lot right in Windows 10 more generally, but the real solution to fixing Windows Phone lies in making the platform more compelling to consumers, and not just at the low end where it’s currently so focused.

For further reading on Windows and Windows Phone:

An archive of all my previous posts from this site on Microsoft is here.

Gearing up for earnings season – new offerings for readers

As regular readers will know, one of the most active times on this blog is earnings season, when I typically write a dozen or so posts on major tech companies’ earnings. You can find past examples here:

I very much plan to do the same again this quarter, starting next week, but I wanted to let you know about two new things I’ll be doing above and beyond these:

Firstly, my regular weekly Techpinions column for Insiders subscribers will be appearing each Monday for the next several weeks, and will provide something of a preview of earnings due that week. So next Monday’s column, for example, will offer a quick preview of things to look out for when companies such as Netflix, Microsoft, IBM and Verizon report earnings later in the week. I’ve highlighted the Techpinions Insiders service before: in essence, for $10 per month (or $100 per year), you get daily articles from analysts including me, often with a data-driven focus, on topics in technology. You can sign up here on the Techpinions site. This subscription will include my weekly posts for the next several weeks, but also gives you access to the back catalog of pieces, including the series I did over the last several weeks on what to look forward to from big consumer tech companies in 2015.

Secondly, I’ll be offering various products off the back of my usual reporting work, which will offer deeper insights into the companies than I can offer in individual blog posts. There will be three tiers available:

  • Overview slide decks (in PDF format) on major tech companies’ financial and operating metrics, updated quarterly after earnings are released, along with my regular quarterly reviews of the US wireless market and US cable, satellite and telecoms providers. This will be available for $10 per month, or $100 per year, as an introductory price for a single user license (multi-user and corporate licenses are also available). More information and subscription signup here.
  • Access to the underlying data behind these decks (in Excel or Numbers format), again updated on a quarterly basis. These will be sold for $500 per company on a one-off basis or $1000 for an annual subscription. I also offer the underlying data behind my US wireless market and cable, satellite and telco analysis for a slightly higher fee. This will allow you to work with the data, create your own charts, run comparisons and so on. Bundled options for multiple companies will be available too.
  • Custom presentations based on the quarterly decks and analysis. Pricing for these presentations will be custom too based on the exact needs, and can focus on single companies or market sectors (e.g. the US wireless market or smartphone ecosystems), and the presentations can be delivered in person in Silicon Valley or via Skype or phone as desired.

You can subscribe to the quarterly decks automatically through a Paypal link here. If you’re interested in any of the other offerings, please contact me (by phone at (408) 744-6244 or by email at jan@jackdawresearch.com). An example of the quarterly decks is below – it’s one I did as part of my Apple profile a few months ago, which was published on Business Insider under the headline “Here’s a giant presentation that tells you everything you need to know about Apple right now”.

Quick Thoughts: Apple’s GoPro

It emerged yesterday that Apple had received a patent on an action camera. The price of GoPro’s stock promptly took a nosedive on the news. Here’s a quick three-point take on all this:

  • Firstly, as others have rightly pointed out, Apple has patented all sorts of stuff over the years which has never materialized a a commercial product, and it’s entirely possible that this particular patent will never materialize in the form described as a product for sale from Apple. It’s most likely that it merely wanted to turn IP acquired from Kodak into a specific patent to both protect its investment.
  • Secondly, I see little value in Apple creating an action camera per se – Apple’s forte is products which combine hardware and software in a tightly integrated way, and most cameras have minimal meaningful software. Action cameras are particularly simple from a software perspective, with all the heavy lifting done on computers after the footage is captured and transferred. I don’t see the value in adding the cost (and potentially weight and bulk) associated with greater computing power to an action camera. As such, I think it’s most likely that Apple will use this technology – it is uses it at all – in future iPhones and iPads, which are of course already among the most popular cameras in the world.
  • Thirdly, I think this says far more about GoPro’s investors than it does about Apple. Yes, Apple has a tendency to disrupt (though not dominate) the markets it enters, and that’s often bad news for the incumbents. But what does it say about GoPro that its investors are this easily spooked? Has GoPro really built so little sustainable differentiation that a mere patent approval concerning Apple can wipe a significant percentage off the value of its shares? And if so, isn’t that – and not Apple’s potential entry – the real problem here?

Quick thoughts: on Apple’s subtle machine learning improvements

John Gruber had a post on Tuesday about the improvements to Siri over time. He was mostly talking about the core functions of Siri itself, but I’ve noticed some other subtler improvements over the past couple of years that I’ve been meaning to write up for ages. Technically, these aren’t part of Siri per se, but they sit in the same category as Cortana or Google Now’s equivalent functionality, so I see them as almost an extension of Siri. What’s striking about these to me is that as far as I can tell Apple hasn’t explicitly talked about them and since they don’t trigger explicit notifications many users might not even notice them.

I grabbed a screenshot of an example of this back in February 2014, and I think it’s the best way to explain the sort of thing I’m talking about:

2014-02-08 07.08.45It’s the first paragraph there that I’m focusing on, and I need to provide some context to enable you to understand what happened here. Early last year I played for a few weeks in a church basketball tournament. I had games regularly each Saturday at a similar time, and at the same location. In my calendar I put simply “Basketball” and never entered the address, since I knew already how to get there. What Apple’s machine learning engine did here was (as far as I can guess 1):

  • Note that I had an item called “Basketball” in my calendar for that morning
  • Make a connection with past appointments on Saturday mornings also called “Basketball”
  • Look up past location behavior in its location database to connect a particular location with past instances of “Basketball” in my calendar
  • Look up this address and calculate driving time between my current location and this destination
  • Present it to me at a relevant time in the Today screen.

Again, Apple has talked up some functionality around using calendar locations explicitly entered in your calendar to provide these sorts of alerts, but I’m not sure it’s ever talked about the deeper machine learning stuff in evidence here. I’ve never seen exactly this sort of extrapolation from past behavior again since this occasion, but I have received other notifications on this screen that it’s time to leave for appointments where I’ve explicitly entered a location in my calendar, based on heavy traffic (it happened to me this past week at CES, for example).

I wonder how long it will be before Apple starts surfacing this information more proactively with notifications rather than merely displaying it reactively when users pop open this screen. And I wonder if Apple will begin talking about this functionality more in future – I note that Microsoft is currently running adds which state that Siri can’t tell you if you need to leave for a meeting based on traffic. That’s technically true – Siri is entirely reactive today – but iOS does tell you if you know where to look. That seems like something Apple could easily fix, but it would turn its machine learning capabilities from a background feature that’s so subtle many users likely miss it into a headline feature. And with that might come concerns about  the data Apple is gathering and how it’s using it. It may have decided that, for now at least, it would rather keep quiet about all this, but I wonder how long that will last.


  1. Another possibility is that Apple merely built a pattern of my past behavior on Saturday mornings without explicitly connecting them to the calendar item. I don’t see that it makes a big difference either way.

Amazon’s changing hardware pricing strategy

I’ve referred to this in passing in several different posts, but I wanted to really devote some time to both researching and writing up a particular trend: Amazon’s changing approach to hardware, and especially the pricing of its hardware.

Kindle: Amazon’s successful hardware product

The Kindle e-reader was Amazon’s first hardware product, and is arguably its only truly successful hardware product too. It has three key characteristics that make it successful:

  • It offers exclusive functionality: it’s the only dedicated e-reader that is designed from the ground up to work with Amazon’s Kindle ebooks.
  • It’s tied exclusively to content sold through Amazon: Kindles are designed for one thing, and one thing only: reading books purchased from the Kindle store. Yes, it’s technically possible to read some free content and content from other sources, but for all intents and purposes Kindles are storefronts for Kindle books and aren’t really useful unless its users are buying those books.
  • It’s competitively priced: although Amazon’s first Kindles cost several hundred dollars, prices quickly dropped, especially as Amazon began selling the devices at, near, or even below cost.

When Amazon debuted the Kindle Fire category, it took in some ways a similar approach, pricing the devices very competitively, at well below what comparable tablets were going for, and the devices were heavily focused on content purchased through Amazon. The exclusive element wasn’t so much in the functionality, however, as in getting that functionality for the price Amazon was charging. But since then, the strategy has diverged further and further from the original strategy that made the Kindle so successful, and Amazon’s other recent devices – the Fire Phone and Fire TV – have arguably continued the pattern.

We never did see those free Kindles promised in 2011

It’s worth looking at the history of pricing for the Kindle and Kindle Fire lines to see how Amazon has handled pricing since the early versions. The chart below shows baseline pricing (WiFi only, no ads) for Kindles since their inception:

Kindle prices over timeIt’s no wonder that there was a slew of articles in late 2011 asking when Kindles would be free, or predicting that they certainly would be shortly (see here, here and here). That was partly just extrapolation from the obvious trend line, but it also seemed to fit with Amazon’s razors-and-razor-blades model for the Kindle. However, what happened in 2012 and beyond was not at all what so many predicted. The baseline price of the Kindle stopped falling in 2012, and the next edition (launched in 2014) actually cost more. The price discount offered on the version with ads masks this a little bit, but the reality is that Kindle prices aren’t falling at all, and in fact they’re rising. If you layer in the other Kindle versions, the baseline price for the more expensive version has actually risen again as well, with the launch of the Kindle Voyage.

Kindle Fires have also been getting more expensive

Now, look at the equivalent chart for the Kindle Fire:

Kindle Fire prices over timeThe trend here is harder to spot, because Amazon hasn’t stuck with a single model throughout the history, rather introducing a series of new models over time. But you can see two distinct trends: individual models have come down in price over time (in the case of the original Fire and the HD 7), while new versions are being introduced at higher prices, actually raising the upper price over time. Amazon has lowered the entry-level price (and you very much get what you pay for at that level) but it’s also moving up the stack into the premium space. The highest base-level price has moved from $200 to $300 to just under $400 over this time.

Let’s go back to the original Kindle model for a second and see how the new line of Kindle Fires stacks up:

  • Exclusive functionality – a key problem for the Kindle Fire line all along has been that they offered no exclusive functionality, only exclusive access to widely-available functionality at a low price
  • Exclusive ties to Amazon content – one of the key problems for the Kindle Fire line is that they could be used for plenty of things that didn’t require another purchase from Amazon. As such, the razor blades and razors model fundamentally didn’t work
  • Competitive pricing – as a result of both of these, Amazon hasn’t been able to keep the prices low, and instead has had to raise prices over time for the most capable tablets in order to truly cover costs.

That last point is the key here: Amazon can’t sell other products on the same basis as it originally sold Kindles because those other products don’t have the same positive effect on other sales as Kindles do. Even Kindles can only be discounted so far before they become loss-making, presumably, even taking into account e-book sales. As a result, both the Kindle itself and Amazon’s other products have been priced increasingly like competing products, at modest to decent gross margins, rather than being sold at cost, because there simply is no significant indirect revenue opportunity from selling these devices. As such, these devices have to be revenue and margin generators in their own right, rather than driving content revenue for Amazon (especially since Amazon has turned video consumption into a Prime perk, rather than a revenue source of its own).

For all these reasons, neither the Fire Phone nor the Fire TV could be priced the way the early Kindles were, because they have to cover their own costs when it comes to revenues. But because many people haven’t understood this shift that’s happened in Amazon’s hardware pricing strategy, they’ve been surprised by the pricing of both devices, which was higher than many expected. Interestingly, the Fire TV stick is the one exception to all this, as the equivalent of those cheap Fire tablets, with more limited functionality at rock-bottom prices. As a result, it’s also sold very well (it’s Amazon’s top-selling electronics device as I write this).

But the implication of all this is that Amazon has lost the two things that made its hardware so compelling in earlier versions, and it’s also lost much of the benefit of selling the devices from the perspective of its other businesses. As such, it should at this point scale back its hardware activities (which are apparently troubled in other ways) and focus on those hardware products which exhibit the same characteristics as its earlier, successful products. That means dramatically paring back its hardware activities, which I’m not convinced the company is ready to do yet. But it’s essential.

Postscript: compare all this to the iPad

It’s interesting to compare all this to the iPad, and what’s happened to its price over the last few years. This is a slightly different chart, in that it shows both the highest and lowest price as well as the average selling price, but it’s comparable:

iPad prices over timeWhat you can see is that Apple has slowly lowered the lowest price of the iPad, partly by keeping older models on sale, and partly by introducing the iPad Mini, such that the entry-level price is now half what it was in 2010, when the device first launched. Notably, though, in contrast to Amazon, it hasn’t done this by creating ever cheaper, crappier tablets, but by keeping very good older tablets in market. It’s also kept the high end of the spectrum remarkably constant, with only a brief blip in 2013 when it introduced a new storage tier. Meanwhile, the average selling price fell for a while and has now stabilized somewhat.

Apple has pursued in some ways the opposite strategy to Amazon, as it often does. It comes in at a relatively high price, giving it the healthy margins it expects while also leaving room for future price discounts. What was so amazing with the iPad, of course, was that it cost far less than people expected even in its original version.