I’ve been chronicling RIM’s death spiral, and much more willing than most to call it as a straight-up suicide by bad planning and management, not just a “bad things happen to good companies” episode. Now comes an open letter from inside RIM describing the unforced errors in excruciating detail.
In other news, the trade press has been abuzz for the last week with stories of a dramatic turn in Apple’s smartphone fortunes – Android supposedly stalling out, with significant gains for Apple from the Verizon iPhone. This report now looks like a classic case study in how to (a) lie with statistics, and (b) get the trade press to inflate a non-story into a nine days’ wonder.
The unnamed author of the RIM letter clearly loves the company. He writes from the perspective of a mid-level executive – someone higher up the org chart than a line manager but not one of the board’s direct reports. There are clues in his perspective on things like UI that suggest to me a marketing person with a bit of technical clue; forced to guess I’d say he’s either a senior marketing guy or a product-line manager with a marketing background.
The letter is candid, fearless, and utterly damning. It calls out the crappy product planning and positioning that I’ve been talking about, and a bunch of other failure modes as well. One of the most interesting is the “run by Canadians” bit – the author says there’s a problem with RIM being too nice and not firing people even for disastrous mistakes. The predictable result is that RIM is clogged with deadwood at all levels.
And yes, the author does have the cojones to tell Lazaridis and Ballsillie, the co-CEOs of RIM, that they have to go. But it’s too late. By the time a letter like this happens, the rot is terminal and the demise of the company not far off. If anything, RIM’s response confirms the indictment. It’s a classic of clueless, defensive, near-meaningless management-speak. I cannot improve on the reaction of this commenter: “Foot. Bullet. BANG.”.
On the other hand, rumors of another demise – that of Android’s astonishingly rapid climb in market-share – turn out to have been (as Mark Twain might have put it) greatly exaggerated. The original claim comes from one Charlie Wolf at Needham & Co.; one of the earlier reports of it was in Business Insider on June 21st. The story was subsequently echoed uncritically by any number of other news outlets, and even led to some serious brow-furrowing on Android fan sites.
I initially ignored Wolf’s claims because the article seemed to me partisan, thinly-sourced and vague. It quickly developed that I was right about the ‘partisan'; according to one of the few critical followups on the story Wolf has a record as one of the biggest Apple cheerleaders on Wall Street and holds a long position in the stock.
But there were clues right up front that Charlie was, er, crying wolf. One is the claim relayed in the Business Insider piece that “in the March quarter…Android’s share in the U.S. fell from 52.4% to 49.5%”. The problem with this is that Android’s U.S. market share hasn’t yet been as high as 50% – comScore reported 34.7% in that quarter and it’s not believable that they were 18 points off.
What was being reported as “share” was, clearly, something else – if it wasn’t simply pulled out of Mr. Wolf’s butt. (This and other reports claimed the analysis was based on figures from IDC.). To confuse matters more, the article contained two graphs, one labeled “U.S. market share” and the other “Worldwide market share”, and while the first one bore no resemblence whatsoever to the comScore and Nielsen trends I’ve been tracking, the second one looked familiar.
Another problem was that we already had comScore’s April figures, and they simply couldn’t be reconciled with any story in which Android lost share in March. In fact comScore indicated a healthy 1.7% Android share growth, exactly on the long-term trend-line fron mid-2010, for the same three-month period in which Wolf was claiming Android had suddenly dropped share.
One of those sets of numbers had to be wrong. My choice was between believing a pro-Apple surprise being retailed by a notorious Apple partisan and a continuation of an 18-month trend being reported by a neutral – not a tough call. Actually the Business Insider article looked like such a mess of vagueness and wishful thinking that I half-thought Wolf’s claims must have been garbled in transmission and that his original analysis was more connected to reality.
Subsequently, an Apple fanboy trying to buttress Wolf’s case pointed me at this report of Nielsen results which claimed that Android dropped a point of share, to 36%, in April. A wildly different figure from Wolf’s 49%, but the fanboy blithely ignored that.
Now come the Nielsen numbers for May. Nielsen says Android share went from 36% to 38%. That’s consistent with the earlier Nielsen report, and if I squint hard I can write off the difference from the comScore numbers as statistical noise. This is the familiar picture of Android rising, with a one-month bobble in the near past that may be just an artifact.
But the May Nielsen report said something else that gave me furiously to think. It said that Android’s share of recent smartphone acquirers had been flat for three months – and when I read that I realized what Charlie Wolf must have done.
That 50% to 49% drop wasn’t “share”, it was some measure of share among recent purchasers. Wolf threw those figures on the table, muddying the distinction, thinking he could con a bunch of business reporters into thinking Android’s share growth had gone into reverse. I think he did this deliberately, but I have a nasty suspicious mind when it comes to Apple fanboys and for the rest of this analysis it won’t matter whether or not there was actual intent to deceive on his part.
The effect was certainly deceptive. Wolf exploited the fact that reporters are in general (a) lazy, and (b) under constant deadline pressure, and thus (c) tend not to question stories that are “too good to check” (like, say, scandalous rumors about Republican politicians). At this point “ZOMG! Sky is falling on teh Android!” is too good to check simply because Android has been riding high for so long that reporters are bored with that narrative line.
And that, boys and girls, is how you inflate statistical flimflam into a news wave. Give reporters something that combines looking slightly contrarian with telling them a dramatic story. They’ll eat it up like candy, and before you know it J. Random Consumer will see the same bullshit tossed at him from a dozen different secondary sources.
Meanwhile, Wolf has left himself plausible deniability. “Well, of course I was talking about recent acquirers! My, my, how my innocent words were garbled in transition!” Then he retires to his fainting couch to contemplate the uptick in AAPL.
But there are substantive questions we shouldn’t toss out with the bathwater here. Is it true that recent acquirers have cooled on Android? Is it a result, as Wolf claims, of the Verizon iPhone? And if so, doesn’t that imply that cumulative Android share growth will stall out in the future?
About recent acquirers: possibly, but the evidence for this is weak. Retrospective surveys of behavior based on what people remember themselves doing (like, what they had for breakfast or what smartphone they purchased recently) are notoriously unreliable. That is, as opposed to watching people eat breakfast or counting the smartphones they actually have in hand.
Such “evidence” is also easily manipulated by changing the definition of ‘recent’ in order to cherry-pick a period in which one competitor’s new purchases were up for exogenous reasons. Another commenter here has discussed the effect of sales of dirt-cheap reconditioned Apple 3GS phones in Q1 – these made Apple very little money but would have inflated its new-user share in a way that is not repeatable.
The truth is, until we know what the time frame of ‘recent’ was, all such figures are basically pretty meaningless.
As to the Verizon iPhone: in late April I wrote:
Even on the very optimistic assumption that Verizon sustains its pace through Q2, Android phones are selling so much faster in aggregate (ratio of about 10:1) that iPhone 4V is barely going to budge the needle on the market share numbers (if that).
And, in fact, the needle has barely budged in the comScore numbers. I think I can see a slight Apple-positive trend since February, but it’s so small that it could easily be statistical noise. And Android, according to comScore and Nielsen, is still gaining share at about 2% a month.
For all the sound and fury around it, the Verizon iPhone has not yet produced any improvement in Apple’s relative market position more dramatic than that in four months.
Of course, it could still happen. Any number of other unlikely and improbable things could happen, too. But usually if a consumer-electronics product is going to make that kind of splash, it happens sooner after initial release. At this point, any realistic hope for an Apple comeback has to be pinned to the late-Fall release of the iPhone 5 or the rumored 4S.
The way to bet, though, is that these will be non-events in exactly the same way as the Verizon iPhone and the iPhone 4 were – hugely anticipated, widely touted as the end of Android’s run, and completely unable to derail it.