the big one: how do you choose speakers? By what features, data?

The Sound.

(To my ears).

And aesthetics.

I have certain things that I value in sound reproduction, hence in the ways speakers can sound. So i’m not chasing numbers (measurements) per se (not that those are inconsequential of course), but rather a sound I want.

When I encounter a speaker that does more of what I’m looking for, at a show, a dealer’s place, another audiophile’s system etc, ...that’s my criteria. I will however be put off if I find the speaker to be ugly. I also want my speakers to be a nice piece of furniture. Fortunately there are so many speakers out there, finding a combination of pleasing sound and aesthetics hasn’t been impossible.

Outside of my own direct encounters with speakers, I will also look to things like speaker reviews, the reports of other audiophiles, to suss out whether people are hearing positive things I care about in a particular model. So for instance all the reports of the textural presence and sonic weight to the sound in the Devore O series speakers ticked some of my boxes, which led me to seek out auditions. And indeed, they ticked those boxes.

Where with my Joseph speakers, I wasn’t led to them via reviews, but by a chance encounter at an audio show. I wouldn’t have necessarily known they stood out for me, unless I’d heard it for myself.

I’ve found my long term CJ Premier 12 monoblocks have been happy to drive speakers big and small, "hard" or "easy" to drive, in my room. So although I favor tube amps, I don’t feel I need to seek out speakers specifically "tube amp friendly."

One thing though is that I can be very intrigued simply by the looks or design of a speaker. It may be the aesthetic beauty, or the audacious or idiosyncratic quality of the design, that makes me feel "Man I’d love to hear what that sounds like." That’s one thing I love about the variety of speakers at audio shows.

@grislybutter These things do not necessarily show up in specs. They are usually design elements that produce a certain effect. Some of it is personal and a matter of experience. For instance, I could care less what a speaker looks like as long as it is well constructed. I want a speaker with sharply limited dispersion, no more than 45 degrees above 250 Hz. Only dipoles and horns meet this requirement. The purpose is to minimize room interaction. I prefer the larger, more realistic sound stage of a line source but in order to work well it has to cover the entire spectrum from 10 Hz to 20 kHz or beyond. This now limits me to very large dipoles. At one point Magneplanar was thinking of making an 8 foot 20.7. The marketing department talked them out of it! In order for a line source to function down to 10 Hz it has either got to be 60 feet tall or run from floor to ceiling, all the way from floor to ceiling. The two most common ceiling sizes are 8 and 9 feet which is why Sound Labs makes 8 and 9 foot speakers. You could easily make ribbons or planar magnetics do this but to my knowledge none do! Fine by me as I prefer ESLs anyway as they can be run without any crossovers to great advantage, excepting one to the subwoofers which I think are mandatory for these speakers to sound their best. As a matter of fact I purchased my Sound Labs without ever having heard or seen them. By design I knew exactly what they are capable of. I had 8 foot Acoustats for decades, a very similar speaker. 

@yxcbandit  The problem with trusting ones own ears is that most of us do not have ears we can trust. How you hear a system depends entirely on your experience listening to a multitude of systems. You cannot know what a system is capable of until you hear it. I built my first amplifier in 1967, a Dynakit Stereo 70. I had no idea how well a system could image until I heard the system of a high school teacher in Miami, FL in 1980. That system had been my target for some 35 years until I managed to build a system that functioned above that level. A large part of it was the room not his system. Back then people were just beginning to deal with room acoustics in residential settings. 

Another problem is your ears can not tell you what to do proactively. My ears do not tell me what equipment to buy. They do tell me whether or not I have that equipment set up correctly as they have that reference built in. 

The key is to gain as much experience as possible listening to other systems. This is not easy. I was fortunate in that I paid my way through school setting up HiFi systems for very wealthy people and worked with the top high end dealer in the area. I got to hear a lot of systems in more intimate environments. It is unfortunate that shows do not work well for this and you have to be deceptive with dealers. In order to get their best service you have to make them think you are qualified (read wealthy) and ready to spend a fortune. Wear gold jewelry and drive up in a 911.

In short, your ears can tell you what is acceptable to you. They can not distinguish  what a system is capable of without having heard it.  

@mijostyn design elements are usually on the spec list but of course it varies. EVERY brand publishes a different set of data

and of course my knowledge is about 3% of yours, my selection criteria is way more basic.

@grislybutter Sometimes too much information can hurt you, especially in this audiophile hobby. 

If you constantly watch speaker reviews, then you'll constantly be fed the same information. This information shall influence your decision more than you'd like to admit. 

Case and point: the Klipsch RP600M was the best thing since sliced bread back in 2019. It got all of the awards from the press and the online community hyped it up to insane levels. I sort of fell into the trap, only because I purchased them without audition at 50% off. The deal was "too good to pass". When I set them up I was underwhelmed. My dad's Q Acoustics 3020 just sound better in every metric (vocal clarity, bass, smoother highs...). 

It's clear that "you get what you pay for". Honestly, most speakers in the entry-level from big brands like Klipsch, B&W, Focal or Triangle perform similarly. It's just that subjectively some brands sound a lot worse to me than others (Klipsch grrrr...). 

Nothing beats a real audition, and with multiple speakers of different brands in the same room. I was honestly shocked how similar the KEF LS50 Meta Wireless and KEF LS60 sounded. The KEF LS60 just have more bass. 

You gotta jump the shark. Go to audition stuff.