The biggest issue I've found is finding projects to work on that are actually useful. Of course you have to start with the basics, but not many people have a use for a simple blinking LED in their life.
Finding projects that you can make that can actually do something that interests you makes a world of difference.
It's like learning programming/OOP through the typical animal examples versus making an app that is something you would actually use or fills a need that you have.
My advice for both Electronics and Programming is to flip it around: find something you'd really like to build and then figure out the tools, skills and a version you can make. Example: My 5-yr-old wanted to build a pinball machine, so we started with using a micro controller to light up an LED and make a noise when a the ball contacted a target. It was a pretty easy intro to the programming, electronics, debouncing a switch (and woodworking!)
My 10-yr old wanted to make a computer game about ancient Egypt and twine was a great way for her to get introduced to programming.
I think musical instruments are a good way to get into DIY electronics, for those interested in both of those things.
Examples include fixing or completely re-wiring the electronics in an electric guitar, fixing or building guitar effects pedals, building midi-actuated devices, building a strobe tuner, adding pickups to acoustic instruments like guitars, cellos, and violins, modding a 3-head tape recorder into a tape delay unit, building pickup pre-amps, winding guitar pickups, fixing or modding old synthesizers or divide-down organs, building new midi interfaces, getting into modular synthesis, building oscillators, filters, sequencers, etc...
Music seems to be the sort of domain where there's always some electronic device you wish you had that would make your life a little bit better in some way, and it often happens that the thing you want is something that no one currently sells, or not for a price you're willing to pay.
Like the "learn scrum by cleaning your apartment" example? Give me a real world example of this working please. If it's worthwhile there must be a case study to build from.
Not sure _useful_ is a good metric, most useful things can be had more easily by buying it. And if its something you really need, then its better to avoid the stress of having to build it right and on time.
But go for something _interesting_, something _you want_.
A music instrument maybe? Automate the blinds in your apartment? Track the motion of earth for long-exposure photography? A multicopter for flying FPV?
Finding projects that you can make that can actually do something that interests you makes a world of difference.
It's like learning programming/OOP through the typical animal examples versus making an app that is something you would actually use or fills a need that you have.