Apr 102013

Right out of the gate, I have a confession to make. I agree with Ed Bott on something. Feel free to stop reading (I would). He thinks the numbers spun by the Gartners and IDCs of the world are fantasyland dart throws. So do I. But why would predictions from analysts that have always supported the Windows feel-good story of global market dominance suddenly fall out of favor with Microsoft pundits like Bott? Because some of their story lines are starting to deviate from narratives that have kept PC advocates smiling contently for years. That must not stand.

Bott’s response to the latest Gartner atrocity – intimating that the Windows share of computers with form factors people want to buy (Ultrabooks, aka MacBook Airs) is contributing to the demise of Windows form factors people don’t want to buy (desktop PCs) – is to take to his ZDNet pulpit and explain why Gartner’s numbers are being misinterpreted and Windows actually looks pretty good in Gartner’s out year of 2017.

Which, true to his arboreal roots, ignores the forest entirely and focuses on the bark of one particular tree.

First, the set-up: here are Gartner’s numbers for computing devices, formatted in the classic 4-year panel of past year, current year, next year and … 4 years from now.

Screeny Shot Apr 10, 2013 9.30.31 AM

Bott laments that pundits have used this information to declare Microsoft’s imminent demise in relevance which, even as written, doesn’t do that. Bott takes particular exception to the 2017 partition between PCs and Ultrabooks, which shows a continuing erosion of the former and a corresponding explosion of the latter.

But even if you take their numbers at face value, you need to actually understand them. With a few exceptions, most of the quick-and-dirty rewrites of Gartner’s press release got the story exactly wrong.

And that’s the second problem. All those reports focused on one shiny thing and ignored everything else in the report. Here, I’ve used my virtual yellow magic marker so you can see Gartner’s data as superficially as all those bloggers did.

Bott proceeds to very helpfully highlight the first row in yellow, an exercise I’ll allow my readers to imagine instead of invoking my right to be a pedantic asshat.

So to Bott, the logical counterpoint is not to make the mistake that every other blogger made and to combine the first 2 categories to show that Microsoft’s outlook isn’t so bad after all. Instead of falling over a -20% cliff in 2017, there’s actually 5% growth! Disaster averted!

As a brief aside, when I first started writing this piece, I was intent on showing that the ultraportable market, now currently dominated by the MacBook Air, is going to seriously bring down Bott’s fist-pumping in 2017. You see, although Apple doesn’t break its sales of products down to the level of granularity in its earnings summaries that show MacBook Air versus MacBook Pro sales, for 2012 it did break out desktops and laptops or “Mac Desktops” and “Mac Portables”; the latest summary only lists total units for both categories. From the 2012 data, if one looks at the average Portable sales of 3.26 million units per quarter, or 13 million per year, and conjectures that Airs made up 1/4 – 1/2 of the sales, that’s between 1/3 and 1/2 of Gartner’s 2012 total for the category. Impressive. But when you look at the 2017 figure of 96 million units shipped, you’d have to live in a fantasy world to think that Apple could sell 24 – 48 million Airs in a year. Fanboy as I am, I don’t think the company is capable of pumping out that kind of volume. After all that math, I was going to scrap the post. Then I realized how inconsequential the first 2 rows of Gartner’s chart were (insert your quip about the general worthlessness of Gartner data here), and saw that which Bott will not name.

Bott may come to the conclusion that PC + Ultrabook sales will mean a 5% gain for Windows in these product categories (at least if we’re talking about units shipped, which is something Gartner/IDC/Microsoft/Everyone Else specializes in), but despite Bott’s assertion, that’s not the important part – and it’s not what is prompting bloggers to proclaim that Microsoft is doomed. Microsoft’s problem is that as a percentage of all connected devices, their share doesn’t dip 20% of a small number, it falls 4% of a much bigger number. Think about the products Microsoft currently has in the Tablet and Mobile Phone categories: Windows RT and Windows Phone. If the hilarious current state of affairs continues, which Redmond is doing a pretty good job maintaining, one could easily defend Microsoft not squeezing 1% out of these markets in 2017. If you assume that Microsoft had about 15% of the connected device market in 2012 by virtue of its IT drone/enterprise desktop phalanx, and one (reluctantly) assumes that it will ship 2/3s of the “Ultraportable” category in 2017, that number drops to 11% of the connected device market.

That’s the picture Bott should be focused on, but I’ve never known a Microsoft shill to be aware of the forest being ablaze when the bark on one particular Acer circinatum is so exquisite.

Update: As if on cue, IDC just announced that PC sales tanked a whopping 13.9% this past quarter compared with 1Q13. But Microsoft totally has nothing to worry about.

Sep 222011

Six months ago, I took apart Gartner’s “analysis” of the tablet market direction, calling particular attention to their absolutely insane predictions for Android tablet performance versus the iPad and the insertion of a 2-year gap in their estimates between 2013 and 2015. I assume this gap was so they wouldn’t have to revise 2 more years worth of “predictions” once the reality of what was happening in the market slapped them upside the head repeatedly.

Gartner’s latest numbers reflect such a reality:

Gartner’s research VP Carolina Milanesi explained the numbers for Android were revised because of “high prices, weak user interface and limited tablet applications”, conditions that will inexplicably reverse themselves by 2015. They also predict a 10% share for Microsoft’s Windows tablets by 2015. Might be time to change the bong water over at Gartner HQ.

Let me offer a possible alternative explanation for the revision: your analysis is constrained by your not knowing what the fuck you’re talking about. Let me also paint a picture of future analyses for anyone thinking of shelling out thousands of dollars for Gartner’s market blindness: every quarter, the only numbers that will be in the ballpark of actuals are the ones reported for quarters past. Gartner will continue to use current market data to incrementally revise numbers for the current year, every quarter, ad infinitum.

I just saved you five grand, so do me a favor and take 5 seconds to click on an ad.

  • RSS
  • Twitter