Another comment on scopes is about 90% of the work I do with a scope is qualitative not quantitative and the spec is usually a quant limit.
So the spec reads something like less than 10% error (usually low) at that freq. So a 10 Hz 10 KHz 10 MHz sine wave at 1 V p-p will display at 1 V p-p on a 20 meg scope. However shove in a 30 MHz 1 V p-p sine wave and (on an analog scope) you'll see a beautiful sine wave at, oh, who knows, maybe only 0.5 V p-p.
I do a lot more quant work with simple watt meters, at least for RF stuff.
A stereotypical example where qualitative accuracy matters not quant is you have a possibly 15 dB amp, and a -25 dB sampler on the input, and a -40 dB sampler at the output and you feed one into the X and one into the Y and shove a varying signal thru the amp and rather optimistically you'll get perfectly smooth lines on the triangle border. The numerical results are irrelevant, whats important is the amp isn't going into compression and what comes out is a perfect linear function of what goes in aka the line is perfectly straight. When you hit compression the end of the line will tilt over and no longer be straight, and thats bad. And your inline watt meter says that happens at 150 watts out or whatever, so I guess you can run cleanly at somewhat less than 150 watts.
Another example is a lot of "is this (four letter words) oscilator oscilating (which would be good) or is this amplifier oscilating (which would be bad)?" and things of that nature where quantative results simple don't matter, you just want to know if there's anything there at all.
Also you can peak up filters and LC networks in general without exact results just a "Turn the screw that way and the sine waves get bigger, turn the other way and output drops" mentality. If you're operating at a couple watts level, then cheap wattmeters will work. Not so much for low power level receiver work, then you use the scope to tune up.
Its a lot easier to use an analog scope at or beyond its limits than a digital scope, but it can be done with digital scopes if you really understand DSP undersampling theory and the sample and hold doesn't aggressively filter higher freqs etc. The reason why pros just dump money to not operate near or above limits while hobbyists can't is a long financial story, my employer can justify a $150K spectrum analyzer in the lab (aka a cheap one) but my wife has issues justifying more than a couple hundred bucks at home. But yeah I prefer the new hyper expensive digital equipment at work and analog at home for that financial reason.
Another fun thing to think about with cheap equipment is I own a 70s/80s era scope but the work I do would only have been cutting edge at a research lab in the 60s, and in the 60s all they had was 60s scopes, so I have it pretty good and could actually allocate my scope money elsewhere. If you want to do 10s (2010s...) work at home its going to be difficult to afford in the 2010s although in the 2030s it might be cheap.