4 ms·

It's well and good to point out that analysis based on supply chain sources may draw inaccurate conclusions and definitely to demonstrate that such analysis has

by erdevs 10y ago

It's well and good to point out that analysis based on supply chain sources may draw inaccurate conclusions and definitely to demonstrate that such analysis has been wrong in the past (it certainly has, for Apple and others). I do not claim to know whether this analysis will bear out in reality or not. I'm merely pointing out that it is nonsensical to blithely dismiss it or claim it may simply be "made up" when it is at least based on primary sources and real journalism. With respect to your specific example, I'd say the WSJ article which the article you linked is referring to appears to have actually borne out with the benefit of hindsight. Q1'13 production supply orders affect supply in subsequent quarters, and a reduction in orders means there is less expected growth. The article you linked expresses skepticism that this is the case, but it appears that it was. iPhone sales growth YoY slowed dramatically subsequent to Q1'13... For example Q1'14 saw ~10% sales growth over Q1'13 versus ~30% YoY growth the year prior and ~100% the year preceding that. To the extent the article indicates that demand and production are falling off from previous growth rates, it appears to have been accurate. The above doesn't mean that this new report will be proven out. Nor even necessarily that the previous reporting you cited was actually correct. But it doesn't appear to be a clear whiff either.

pg314 10y ago
You are being very charitable to the WSJ. The WSJ article in question said in the second paragraph [1]: Apple's orders for iPhone 5 screens for the first quarter, for example, have dropped to roughly half of what the company had planned to order, the people said. In the three first months of 2014 Apple sold 37.4 million iPhones, compared to 35.1 million in the three first months of 2013 [2]. Unless you believe that they internally projected to sell anywhere near 75 million phones (more than the 47.8 million iPhones sold in the last three months of 2013, traditionally their strongest quarter by far), that rumor reported by the WSJ was flat out wrong. [1] http://www.wsj.com/articles/SB10001424127887324235104578241051730364998 http://www.wsj.com/articles/SB100014241278873242351045782410... [2] http://www.apple.com/pr/library/2013/04/23Apple-Reports-Second-Quarter-Results.html http://www.apple.com/pr/library/2013/04/23Apple-Reports-Seco...
- erdevs 10y ago
  I hear what you're saying, but I am not sure you're interpreting the supply chain order implications correctly. You are assuming a 1 to 1 ratio of expected vs actual component supply order volumes and subsequent net sales. That is not at all how it typically works, and while I agree the WSJ could have and should have been more clear on this: the point of the article wasn't that the iPhone would sell half as much as Apple had forecast. I can see why you and other reasonable people would draw this conclusion from the article (hence my criticism that they could and should have been more clear), but it isn't implied. Let's say that for Q4'12 Apple ordered 41M screens (vs say 30M in Q4'11), tracking toward a roughly ~35% YoY growth rate at the time. And, as per usual practice, so as to ensure their component partners in the supply chain are adequately tooled and ramped up for the future, they set expectations for Q1'13 and beyond, but at slightly lower growth levels than Q4'12's order because they are savvy and realize this is a maturing market where growth naturally slows over time. Without obligating themselves to actually order these future quantities in Q1'13 and beyond, they place contingent forecast orders representing say a ~28% YoY growth rate estimate for Q1'13 and beyond. Say that means they communicate a forecast order of 44M screens for Q1'13. Now, as Q4'12 rolls along and Q1'13 component orders must be finalized and committed, Apple sees actual sales growth might be on pace for ~18% YoY. It looks like they will sell ~35M iPhones in Q1'13 vs the ~41M screens they have on hand from Q4'12's order, so they'll be left with a 6M screen inventory overhang from Q4'12's order. (I'm oversimplifying and assuming single quarter component turnaround time here for sake of illustration). They see that growth is tapering off going forward too (tracking down over time toward the actual ~10% YoY growth you cited by Q1'14). Q4'12's sales landed slightly slower than anticipated when they placed their order for it in Q3'12 as well, in line with this slower growth trend, and so they also have an overhang of, say, 4M screens from Q3'12's order for a total inventory overhang of 8M+4M = 12M screens by the end of Q4'12. They deduct this overhang from their projected Q1'13 screen order, as well as lowering their estimate from needing 44M screens in Q1'13 to needing, say, 36M screens total for Q2'13 As such, for Q1'13 they order 36M needed screens minus 12M screens in overhung inventory, or 24M new screens. So, Q1'13's order goes from a projected 44M screens to 24M. In this scenario, a screen component supplier might say Apple "cut their Q1'13 order in half". But Apple's expected sales for Q1'13 and beyond wouldn't have been off by 50%, but more like less than 20% (41M vs 35M). Also bear in mind that Apple's internal forecast may not have even been off by this much. Companies sometimes slightly overprovision in periods of uncertainty if they know they can sell through the overhung supply later and simply dial back subsequent supply orders while the overhang is cleared through. This may be exactly what Apple did at this time around the iPhone 5's sales cycle, so there internal forecast may have been off by only 10%... Or not at all. Yet, it would still be true from a component supplier's perspective that their order for Q1'13 was cut roughly in half. This is all grossly oversimplified and entirely hypothetical (though based very roughly on the volume figured you cited). Yet it illustrates the point that there certainly needn't be a 1:1 correlation between supply order changes and actual sellthrough vs forecast. This is only more true in the real world where myriad complexities and order terms can dramatically affect supply chain orders without implying a horrible misforecast by the buyer or any dramatic or unexpected sales trend. This would make what Tim Cook told investors absolutely correct-- don't overextrapolate from (potentially faulty) supply chain sources. But it would also make the WSJ article perfectly accurate... And nowhere did the article state that Apple's estimates were 50% off nor that investors did or should expect a decrease in demand or even a drastic decrease in growth. So, they were both (potentially) right. Point being, it is faulty to assume a simple 1:1 correlation between supply chain order changes and resultant actual sales vs forecasts, as your post does.

by erdevs 10y ago

pg314 10y ago

You are being very charitable to the WSJ. The WSJ article in question said in the second paragraph [1]: Apple's orders for iPhone 5 screens for the first quarter, for example, have dropped to roughly half of what the company had planned to order, the people said. In the three first months of 2014 Apple sold 37.4 million iPhones, compared to 35.1 million in the three first months of 2013 [2]. Unless you believe that they internally projected to sell anywhere near 75 million phones (more than the 47.8 million iPhones sold in the last three months of 2013, traditionally their strongest quarter by far), that rumor reported by the WSJ was flat out wrong. [1] http://www.wsj.com/articles/SB10001424127887324235104578241051730364998 http://www.wsj.com/articles/SB100014241278873242351045782410... [2] http://www.apple.com/pr/library/2013/04/23Apple-Reports-Second-Quarter-Results.html http://www.apple.com/pr/library/2013/04/23Apple-Reports-Seco...

erdevs 10y ago
I hear what you're saying, but I am not sure you're interpreting the supply chain order implications correctly. You are assuming a 1 to 1 ratio of expected vs actual component supply order volumes and subsequent net sales. That is not at all how it typically works, and while I agree the WSJ could have and should have been more clear on this: the point of the article wasn't that the iPhone would sell half as much as Apple had forecast. I can see why you and other reasonable people would draw this conclusion from the article (hence my criticism that they could and should have been more clear), but it isn't implied. Let's say that for Q4'12 Apple ordered 41M screens (vs say 30M in Q4'11), tracking toward a roughly ~35% YoY growth rate at the time. And, as per usual practice, so as to ensure their component partners in the supply chain are adequately tooled and ramped up for the future, they set expectations for Q1'13 and beyond, but at slightly lower growth levels than Q4'12's order because they are savvy and realize this is a maturing market where growth naturally slows over time. Without obligating themselves to actually order these future quantities in Q1'13 and beyond, they place contingent forecast orders representing say a ~28% YoY growth rate estimate for Q1'13 and beyond. Say that means they communicate a forecast order of 44M screens for Q1'13. Now, as Q4'12 rolls along and Q1'13 component orders must be finalized and committed, Apple sees actual sales growth might be on pace for ~18% YoY. It looks like they will sell ~35M iPhones in Q1'13 vs the ~41M screens they have on hand from Q4'12's order, so they'll be left with a 6M screen inventory overhang from Q4'12's order. (I'm oversimplifying and assuming single quarter component turnaround time here for sake of illustration). They see that growth is tapering off going forward too (tracking down over time toward the actual ~10% YoY growth you cited by Q1'14). Q4'12's sales landed slightly slower than anticipated when they placed their order for it in Q3'12 as well, in line with this slower growth trend, and so they also have an overhang of, say, 4M screens from Q3'12's order for a total inventory overhang of 8M+4M = 12M screens by the end of Q4'12. They deduct this overhang from their projected Q1'13 screen order, as well as lowering their estimate from needing 44M screens in Q1'13 to needing, say, 36M screens total for Q2'13 As such, for Q1'13 they order 36M needed screens minus 12M screens in overhung inventory, or 24M new screens. So, Q1'13's order goes from a projected 44M screens to 24M. In this scenario, a screen component supplier might say Apple "cut their Q1'13 order in half". But Apple's expected sales for Q1'13 and beyond wouldn't have been off by 50%, but more like less than 20% (41M vs 35M). Also bear in mind that Apple's internal forecast may not have even been off by this much. Companies sometimes slightly overprovision in periods of uncertainty if they know they can sell through the overhung supply later and simply dial back subsequent supply orders while the overhang is cleared through. This may be exactly what Apple did at this time around the iPhone 5's sales cycle, so there internal forecast may have been off by only 10%... Or not at all. Yet, it would still be true from a component supplier's perspective that their order for Q1'13 was cut roughly in half. This is all grossly oversimplified and entirely hypothetical (though based very roughly on the volume figured you cited). Yet it illustrates the point that there certainly needn't be a 1:1 correlation between supply order changes and actual sellthrough vs forecast. This is only more true in the real world where myriad complexities and order terms can dramatically affect supply chain orders without implying a horrible misforecast by the buyer or any dramatic or unexpected sales trend. This would make what Tim Cook told investors absolutely correct-- don't overextrapolate from (potentially faulty) supply chain sources. But it would also make the WSJ article perfectly accurate... And nowhere did the article state that Apple's estimates were 50% off nor that investors did or should expect a decrease in demand or even a drastic decrease in growth. So, they were both (potentially) right. Point being, it is faulty to assume a simple 1:1 correlation between supply chain order changes and resultant actual sales vs forecasts, as your post does.