What Apple can learn is to manage expectations. Pebble lost credibility because they promised a lot and then could not deliver.
I don't agree with this narrative and I never have.
When your Kickstarter goes viral and you score a 100x multiple on your modest proposal—a proposal which involves custom electronics hardware—you face a monumental problem: having to put a one year warranty on a beta++ prototype you shepherded into mass production with a virgin organization in building mode. You're not exactly a six-sigma shop that never gets a simple ignition switch wrong. The American government isn't going to step in to tide you over if an unexpected recall of 2.5 million vehicles tips your balance sheet into the red.
I bought two Pebbles, and presently one is dead and the other has an issue with garbling the display where I can't always read what it says, and sometimes when I can read what it says, the time isn't correct, because it hasn't successfully updated for 10 minutes.
The dead watch (which went from functioning perfectly to hardly functioning at all in the space of a day) is a more advanced case of the same problem. It still seems to function in many respects, but the screen just flashes garbage with here and there a surprise appearance of a proper image. At this point, the correct screens are so rare it's impossible to navigate any watch menu (unless I use my other watch as a guide dog for the blind).
On the one that still barely works, some of my watch faces are more garbled than others. I've done plenty of micro-electronics troubleshooting in my time, and this looks a lot like loss of signal integrity between the CPU and the display chip due to an insufficient voltage margin. It could be as simple as the battery not responding to current transients as quickly as it once did, so that you see a bounce brown-out that barely dips enough to induce a few corrupt bits on a data signal. You're oh so close. Close, but no cigar. It only takes one wrong bit per screen update to ruin everything, if it's the wrong wrong bit.
Back when People was slipping schedule, people complained bitterly that other companies manage to bring products to mass production a lot faster than Pebble. These people are idiots. If Pebble experiences a 20% failure rate in the field of their delivered product, and has to honour it's warranty on all these failed watches, what exactly is Plan B? Fold up their tent? As it turns out, they did raise some venture capital. VCs don't come along with a fistful of dollars to throw into an RMA pit.
Given their economic and technical parameters, I thought Pebble did just fine in delivering what they did deliver. After a certain point, though, I actively disliked their communication policy. They handled the issue of the colour additives messing with their plastics rather atrociously.
I made one attempt to get my failed Pebble replaced and the communication dried up on their end. By the time I figured out I wasn't going to get the next response, I had become too busy in my own life to pursue it, and I was beginning to notice my other watch glitching and becoming erratic. I figured I would wait until it was bad enough to declare it officially failed, and pursue replacement of both watches in tandem. Based on when I received my watches, I'm guessing I have about a month left on my warranty now.
If they honour their warranty (and the replacements prove more reliable than the first two) I'm still a big believer in the Pebble platform. I couldn't do what I wanted to do with these watches until the 2.0 SDK came out and that has only been out since my watches became unreliable. No point investing in the new API until I find out whether I'm going to have working watches or not.
No matter what Apple invents, it's simply not going to be as open as what I'm willing to invest my time into.
If Pebble turns this into a seventeen email exchange with a week-long delay after every volley, I'm just going to walk away from my Pebble investment, like any other investor who dabbles in a risk bearing portfolio.
Pebble's original delivery problem was inherent in how Kickstarter works. There is no such thing as an electronics production plan where 1x, 10x, and 100x production volumes share identical schedule lengths. Every order of magnitude is a different fish. Sometimes a volume bump permits you to leap over roadblocks. Sometimes the volume bump makes it harder to sleep at night, out of fear that the whole thing turns into a GM ignition switch.
Because Kickstarter feeds on momentum, no adult in the room ever shows up to point out that the subscription euphoria is putting the project onto a negative delivery ramp: that addition subscribers are having the net effect of delaying the product for everyone.
Pebble could have afforded to have a 30% return rate on a production run of 1000 units if a subsequent production run was already well subscribed and they had by then ramped up their production process to a much higher success rate.
They certainly couldn't afford a garish return rate on 80,000 units all delivered at once.
Scenario A: Somebody hands you the keys to a $10,000 econobox and asks you to back it out of a long, tight driveway. You're a little nervous, and exercise more caution than normal.
Scenario B: Somebody hands you the keys to a $100,000 Range Rover and asks you to back it out of a long, tight driveway. You're a little nervous, but you take it extremely carefully and all goes well.
Scenario C: Somebody hands you the keys to a $1,000,000 Ferrari and asks you to back it out of a long, tight driveway. You're more than a little nervous. The first time you tap the gas pedal, it jumps back six feet. You attempt to shift it into drive to pull forward and start again, and something makes a thunking sound. Is it supposed to sound like that, or did you just trash the tranny? You're not so familiar with this kind of car.
How long is it going to take you to back the Ferrari out of the drive way? Half as much time because the car is 10x more powerful?
Sorry, Kickstarter idiots, life doesn't work that way. Yet it seems to be what most unthinking subscribers presumed in rallying behind the myth that Pebble over-promised and under-delivered.
I just wish my two watches worked reliably enough to be worth rolling up my sleeves and custom programming them as I originally envisioned back when I first subscribed.