How do you get current customers to upgrade cameras? How do you get someone new to buy a dedicated camera for the first time?
Answer those two questions, and you have a job in Tokyo.
The industry has mostly punted or given lip service to the second question (finding new users). That's because they can't respond to the yearly smartphone camera improvements as the ROI isn't there to turn processing chips that fast for cameras. Thus, the camera companies are now in an existential fight that will require them to keep as many existing users upgrading (or switching systems) as possible.
Worse still, some of the long-established product iteration keystones are no longer working to generate upgrades from existing users (better sensors, more pixels, faster frame rates, etc.). A half stop dynamic range boost and 20% more resolution is simply not dramatic enough to get most sane folk to fork out another US$3000 for a new camera body. Put in an automotive context, that's like offering 1 MPG more and a 0-60 speed reduced by a half second and expecting that to generate a sale on its own. Not the reason I'd buy a new vehicle when my current one is perfectly operational.
Cameras now have to wear out or be dropped/broken in order to get someone to rationally upgrade now.
This is why I continue to reiterate the same thing that drove my multi-decade Silicon Valley career: what are the customer pain points? What user problem has to be solved in order to get another grab at the customer's wallet?
Sony sort of accidentally hit on one such thing while doing their usual product iteration: tracking focus. I've been observing for awhile that the way that most users approach "focus" isn't the way the camera makers were trying to make them do it. The dissonance in that made for a lot of focus failures, even on cameras capable of doing precise, fast focus. Why? Because the user had to learn something in order to make focus work. Worse still, the user had to change what they were doing if the situation in front of them changed. In other words, there was another thing for them to control.
What I'm noticing a lot of Sony users do (and Nikon users, too, with the D5 generation DSLRs), is what I'd call identify-and-track. "Hey camera, this is what I want in focus, now follow it." Nikon users will recognize that as AF-C with 3D Tracking Area mode. Sony users will recognize that as some variation of Flexible Spot. Put the cursor on the thing you want to track, press the AF-ON/half-press the shutter button, and re-compose or follow the subject as it moves in the frame. The result for most people is that the camera now seems to do what they want it to do (and this is why Nikon's convoluted 3D Tracking function on the Z cameras is wrong: the user has to spend too much time fiddling with and controlling it).
The next inevitable user desire, whether they know it or not, is for the camera to use AI or some algorithmic way to figure out what should be tracked in the first place. At that point, the user doesn't have to say "hey camera" at all ;~). Maybe they'll have to say "no, not that," though ;~)). This is why Face Detect and Eye Detect got so much attention, by the way: if you want the camera to focus on a human, then those are good choices to look for. (Note: Olympus thinks you want to follow trains, planes, and automobiles. Nope, not faces, not eyes, not trains, not planes...but all of the above, automatically figure out the subject and follow appropriately.)
As I've tried to point out for over a decade, the other big pain point is workflow. Yes, that means that some people want to immediately have something shared, without having to pull out another device and doing any work on it. But they also want things backed up automatically (where's the Time Machine equivalent for cameras? iPhones have Apple iCloud Photos, after all. Funny thing is, Nikon has Nikon Image Space and Canon just shut down their cloud approach. The problem isn't the cloud, it's how you use it and don't require the user to do anything).
The list of things that the camera could help with downstream of actually taking the photo is actually pretty extensive, but nothing regarding workflow is being done by camera makers. That's software, after all, and the Japanese camera makers think they only make and sell hardware. No, they sell useful (or non-useful) products, and these days in tech, that means software that makes hardware solve problems, do chores, make decisions.
Mirrorless had a bit of a selling wave because of another coincidental thing: it solved a pain point and user problem, not that the camera makers immediately recognized and marketed that (they are now). With film cameras we waiting hours to days to see what our shot looked like. With digital we all started chimping, and that reduced the wait down to seconds and minutes. With mirrorless, we see the preview of what it will look like before we take it, and can adjust prior to the moment happening. With mirrorless, correctly exposed and composed selfies are even possible ;~).
But frankly, that's not enough, and particularly for high-end practitioners. What we see in the EVF on a mirrorless camera isn't actually our raw data. Histogram and Highlight displays lie to us about what's actually going on in the data. Why we don't have a different measurement system for raw shooters I don't know. I've asked for it for 17 years now. So have plenty of other high-end shooters. Talk about not solving a user problem.
Then there's the half-done features our cameras seem to get. Sony's pixel-shift, for instance. Okay, we get a huge batch of data, but no actual image. To create a useful image requires a lot of downstream work (what did I say about workflow?), and we have no way to evaluate—via preview or chimping ;~)—what it will look like. Nikon's Focus Shift [sic] feature is worse: not only do we not see the result (though the latest cameras can give us a geeky estimate of what's in focus after the fact), but we have no idea what the values we enter into the system actually do.
The problem I keep coming back to is that the camera makers keep throwing more features into cameras as if that's the problem with cameras not selling. Not that there aren't features I wouldn't want to see added, but the real reason I ever want a feature is because it solves a problem for me. So it had better actually solve that problem, not just tease me.
Camera makers need to spend much more time solving our problems than they do on iterating the tenth time for a feature few use (let alone understand; what's Regist. Face Priority again and when would I use it? Does anyone use it?).
Everyone reading this probably has a very competent camera (that would be pretty much anything post 2012, when camera sales peaked). You might consider upgrading to a new camera, but it's amazing to me how many of you are basically stalled at the moment. You're stalled because the camera companies aren't necessarily changing the things you think would do you the most good.
In the Nikon world, for instance, you'll find D6xx, D7xx, and even some D8xx users who would consider moving to full frame mirrorless. In the Nikon lineup, though, Nikon put the Z6 and Z7 in the middle. The Z7 is not > than the D850 or even =, it's < than. This creates a buying friction that keeps many from pulling the trigger. A D800 user thinking to upgrade sees the D850 as the best choice, but then wonders what happens when DSLRs go away. (Marketing message, Nikon. Get one.)
Meanwhile, Canon came out with the RP and R, which really are older generation sensors in lower end to middle market cameras when you look to what DSLR they equate to. So Nikon shot middle, Canon shot lower. And then they wonder why you're not buying.
Both Canon and Nikon seem to be a little bit too worried about preserving any remaining high-end DSLR volume than they are in moving you to mirrorless. This was exactly the issue I worried about prior to these two companies' inevitable transition from DSLRs to mirrorless: too protective of the existing product. They should have blown away their existing products and gotten everyone to transition.
Sony, on the other hand, has had five years to figure things out, having done this same transition (almost as poorly) quite some time ago. To their credit, Sony has done some very good things in that time. Yet, when I contemplate the A7R Mark IV versus the Mark III, I'm down to "does 20% more resolution solve any problem for me?" Not particularly, and it introduces some new problems (file size, lens choice, diffraction avoidance, etc.). So things really start to boil down to mostly some changes in ergonomics. You might have noticed that I stuck with the A7R Mark III for my current "best-all around camera" (number two behind the D850). And camera makers wonder why the high-end isn't selling as well as they guessed it would.
The camera makers will eventually figure things out and the market will stabilize, or they'll pilot the plane right into the ground. We seem to be a ways from either option happening (unfortunately and fortunately, respectively ;~).
What I worry about today is peripheral damage. Can the camera shop I rely on stay in business? How about all those software makers trying to unseat Adobe? Will the flash and tripod and accessory makers other than the Chinese knockoffs of IP still remain standing?
I think we're about to find out.
