There was a lot of confusion yesterday when Verizon’s results were discussed, with more than one commentator confusing activations and sales. For the record, what Verizon announced was that 51 percent of its activations were iPhone, not 51 percent of its phone sales.

If you doubt the importance of this distinction, I have one word for you: T-Mobile. As of 11th April, the carrier had two million iPhone activations. Its iPhone sales as of the same date? Zero: T-Mobile didn’t start selling iPhones until the following day.

The difference between the two numbers is particularly dramatic with high-end handsets like the iPhone … 

A low-end handset may well be discarded when its owner upgrades. They buy the replacement: one new sale for the manufacturer, one new activation for the carrier. But iPhones typically remain useful for several generations. There are plenty of people out there still happily using an iPhone 4.

When someone upgrades to their shiny new iPhone 5s, the chances are their iPhone 5 is sold on the used market or passed on to a family member or friend. Perhaps that person passes their old iPhone on to someone else. One new sale for Apple, two or more activations for the carrier(s). Those activations Verizon cites will for sure include a great many older iPhones not sold by the carrier.

This kind of confusion is not unusual. When it comes to smartphone numbers, we frequently see sales numbers of one handset compared to shipment numbers for another, as if they were the same thing.


The same applies to all those supply-chain rumors, telling us that the iPhone or iPad sales are rising or falling based on component production numbers at a single supplier. As Tim Cook has cautioned:

I’d stress that even if a particular data point were factual, it would be impossible to interpret for our overall business. Yields can vary. Supplier performance can vary. There’s an inordinately long list of things that would make any single data point not a great proxy for what’s going on.

Anything other than an official statement from Apple relies on proxies: other stats that provide clues on sales numbers rather than evidence for them. That’s why we’ve seen wildly-varying estimates of the relative sales of the iPhone 5s and 5c.

Look at mobile and web app connections, and you get three-to-one in favor of the 5s. Look at claimed production numbers, and you get 50:50. Ask consumers what they own and you get two-to-one.

Of course, when a company reveals as little information as Apple, we can’t escape from the guesstimates. It’s just important to understand what the numbers really are, and what they do and don’t really mean. That’s the context we aim to provide here at 9to5.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

21 Responses to “Why iPhone longevity means iOS carrier activation share doesn’t resemble sales”

  1. Still, I thought Verizon could only activate their special versions of iPhone models due to CDMA technology…


    • This is true, but the point is that if I bought a verizon iPhone 5 in July, then give it to my sister when I bought a 5S last month and she activates it on her number, it counts as 3 activations of iPhones, but only 2 sales of iPhones, which is why activations as a metric of sales success is misleading.


      • Robert Dupuy says:


        However, then I have to say sales as a measure of installed base is misleading.

        If you sold 10 Android devices and 9 of them are now in a landfill – I’m not going to count those in the landfill as part of the potential market for my app.

        I mean it’s all good discussion, but activations is still the number I want. Apple wants to know the sales – lucky for them, they do know the sales.


    • standardpull says:

      Not exactly. Any iPhone 4S, 5, or 5S can be activated on Verizon. But it may be more or less functional based on the sub-version of the device. It isn’t to do with CDMA, but with the bands in use (the same bands are used by both CDMA and GSM).

      Regardless, one device is generally activated one or more times over its lifetime.


  2. Then there are instances such as mine, wherein I bought an iPhone with a relative’s upgrade price, switched it to my name, unlocked it, jumped to T-mobile. AT&T gets a sale & activation number, T-mobile only gets the activation number. Lies, Damn Lies & Statistics folks.


  3. rogifan says:

    I’m not sure what the point of this article is. Verizon hasn’t changed the way it’s reporting iPhone figures. They’ve reported activations in the past and they’re doing so now.


    • Ben Lovejoy says:

      That’s correct, but a lot of commentators are confusing the reported activations with sales


      • rogifan says:

        True but is that any different than any other quarter? To me this would only be meaningful if the thought was the numbers this quarter aren’t comparable to previous quarters. For example, if a smaller percentage of activations this quarter were new sales as compared to previous quarters or previous year quarter. Otherwise to me it’s like the whole shipped vs sold argument. As long as you’re comparing Apples to Apples (pardon the pun) it doesn’t matter what you call it.


      • Isitjustme says:

        So google activation of androids are not sales too?


    • Paopao Wudi says:

      I think the point of this article is to notice you, evaluating the phone carrier’s performance from those data, not apple’s. if you buy stock, don’t make wrong decision.


    • standardpull says:

      The point is: make sure you’re comparing apples to apples. Sales are not Shipments; activations are not sales. If you get different data from different reports they very well may not be comparable.

      iPhone 5S activations will MOSTLY be new devices (right now, as people are unlikely to be RE-activating brand new devices.)

      iPhone 4 activations now will MOSTLY be re-activations, as people are unlikely to be firing up such an old device for the first time right now.


  4. Jim Phong says:

    A lot of nonsense babbling here…
    So no one sold any iPhone in the past years and just 10-12 people worldwide bought an iPhone that got re-activated 300-400million times thru the years, uh ?


  5. The writer said: “Of course, when a company reveals as little information as Apple, we can’t escape from the guesstimates.”

    Name one company that gives out as much smartphone information as Apple…


  6. “Of course, when a company reveals as little information as Apple”

    Oh really, what were Google and Amazon’s device sales figures again? Oh that’s right, they don’t break them out AT ALL.

    It’s absurd to chide Apple about revealing information when the are the most transparent of all mobile manufacturers.


  7. barrysayer says:

    Networks will release figures that put them in a good light. The longevity argument is a bit misleading too where 1 new device causes a domino of 3 or so activations.. There’s a very good chance that all 3 ‘new’ activations are also either 3 disconnections or upgrades, in other words, churn. There will of course be churn off one network to a different network but that’s always been the case.


  8. I think this also influences the whole ‘market share’ vs ‘usage share’ argument – that Android vendors may be pumping out more handsets (and selling them), but when you look at usage metrics, not only do iOS users stay more engaged with their devices, but there are more ‘hand-me-down’ iOS devices out there, meaning people get more value and longevity (and subsequently, iOS is used more).