You need to calibrate only the screen where you do your color editing.

In addition, you may want to calibrate your other screens where you watch your products.

Any errors on any input devices do not matter anymore after you have done your editing.

Any errors on any other output devices do not matter when they belong to other people. People are used to watch colors with these errors on their devices (or they may calibrate them).

If it matters that much to you, then calibrate each tool.

You need a colorimeter that has software support for each device.
I only use one on my PC, to be sure my edits are accurate. The rest of my devices are as they are.

I stick to just one device for editing. How they look on various electronic devices worries me far less than how they look when printed in books and magazines.

This has nothing to do with the Apple ecosystem. This is just a fact of manufacturing because different screens will have different qualities. You will never have full control over this aspect because you cannot control what screens your audience will use.

But that being said, you can minimise the effect, and that is with a proper colour-managed workflow. To have a proper colour-managed workflow, there are few things to understand. Your camera, lens and the lighting conditions will all have a massive effect on your colour reproduction. This is why post-production usually is done in two parts, primary colour grading and secondary colour grading. Primary color grading refers to the process of making the output of a camera look technologically correct, while secondary color grading is usually done for creative effect. There is a right way to do primary color grading, but there is no right way to do secondary color grading. Now this in more practical terms means that in primary color grading you first of all focus on making the camera look like real life and in the second step you take that “real life” corrected image and turn it into something aesthetically pleasing.

Now to achieve this in more practical terms, you need to do a few different things. Firstly, you need a corrected monitor. That, you achieve by buying a so-called colourimeter, which is just an image sensor behind a lens, which the manufacturer has actually calibrated. When a manufacturer says it has factory calibrated a monitor, it means it has been generally colour corrected, not your monitor specifically, because there are also variances between copies of the same monitor model. So, now you need to download the software called DisplayCal, which will then show different colours on the monitor. The colorimator will take an image of the colour that your monitor has shown and calculate the difference between these two. Depending on the number of steps, it does this a few thousand times and calculates a big table with correction values, such that in the software, wrong values will be sent to the monitor And with the natural offset your monitor has, it will now show up correctly to your eye. This step takes care of your monitor.

Now, as for your camera, that is a little bit more complex. For the full details, I think you should read the darktable manual. Even if you don’t use the programme, it will give you a good basic understanding of what’s going on with your digital image. So, firstly, you need to understand that no digital camera actually sees colour. It just sees various intensities of the red, green and blue channels. And from there, a mathematical model is used to basically blur the image together in an elaborate way, which creates colour. Now, basically before starting out you’ve already lost there, furthermore your lens and the lighting conditions will have a massive influence on how colour is rendered in the final image. Therefore you need something which has known colour beforehand. This is done by using a colour chart, basically a sheet of plastic or paper which has been created with a high-end digital printing process where the colours are perfectly exact. You should use Darktable, which has the by far easiest colour workflow that I’ve ever worked with, to correct the colours with the so-called colour calibration module. It really is that simple: Just align the patches in the software with the patches of the actual color chart and there you go. Press a button and there’s your finished colour profile. Now, there are obviously more elaborate things going on under the bonnet, but for you, the end user, it is being kept dead simple. And that’s what makes it truly elegant.

To guarantee maximum compatibility for your finished images, you should always export as sRGB, because that profile has been around for so many years, and every device, no matter how bad the monitor is, supports it.

And with that, you basically have done everything you can to guarantee colour consistency between devices for your audience. To recapitulate, firstly, you should use DisplayCal with the colourimator to profile your monitor. Therefore, you know it is colour neutral. Secondly, if you’re shooting under a certain light source just take a picture of your color chart under this light source or just in the room you’re currently shooting and later use darktable to edit your images because it has the simplest colour management workflow that I have ever seen. Lastly, export your images in the most compatible colour format, which is sRGB. Because every device supports it, it has been around for so many years, no matter how bad the device is, it can display sRGB. This will guarantee the most colour consistency. Do not worry about HDR or any or other of these bits and bobs. Just do sRGB and JPEGs and you’ll be golden.

Edit: Fixed spelling mistakes and clarified some others.