Although the answer to this question would seem to be simple, that being "sound", I wonder if other factors enter the equation, such as ease of use, aesthetic appeal, compatibility with hosts, customer service/support, even word of mouth on this forum, or perhaps all.
Personally, I always go for sound first. And also, the acid test is if those sounds get me thinking creatively. Many is a time when I've played a keyboard in a store and come up with so many ideas that it prompted me to buy the product. Obviously, it's a little harder for software synths/libraries to cater to this kind of shopping ideology, but when I hear demos on line that blow me away or else have a couple demo patches to play with, that usually does it for me.
The reason I bring this up is that there are several developers at this forum who are very passionate about what they do. And I think that's great. the more blood, sweat and tears (at times) they put into their product, the better off we the customers are. Because of this dedication, tempers can run high and people get defensive when their product is dissed. But I honestly wonder if vehement criticism by some customers really taints the potential purchase by another. I know we all ask for opinions around here and other forums before buying, but does it really have a big impact on the sales?
I'm getting a little bored with the amount of bloodbaths that come out of debates between which developers sounds are better. I do think articulating one's opinions better on the web would help to clear up many misgivings too. Just imagine if everyone said:
"I don't find the sounds from ABC lib to be useful for my application or musical style" as opposed to
"Developer ABC sucks".
Obviously the first example takes a little more time to type out and perhaps think of in the verbiage but it certainly would help avoid such heated arguments.
As far as I can tell, every developer's product has its merits. I mean, does anyone set out to work hard on a crappy product? Heck no!
As for the tempers of the customers, well, I can understand their point. They drop a lot of money in some cases on a lib only to find out it is buggy or wasn't exactly what they expected. I own Gigastudio Orchestra, Miroslav Philharmonik, and MOTU Symphonic Instrument and none of the three get any use (giga is uninstalled on my PC actually) because they didn't do much for me. On the other hand, I'm enthralled with my East West libs, VSL sounds, Korg Digital Legacy Collection, Arturia CS80V, and most recently Gary Garritan's Big Band plug-in. Couldn't be happier.
And as for the personalities involved, I find it hard to dismiss ANY of the developers here or on other forums. I admire the fact that Gary Garritan has created such a helpful online community (as well as create an incredibly effective performance interface for his sounds!), Project SAM's awesome brass instruments and great customer support (my initial Solo Sessions DVDs didn't read in my DVD player so they rushed me a replacment copy), Herb/Dietz's amazing musicality that courtesy, Sonic Implant's fantasic crew, and yes, Doug and Nick from East West who have been nothing but responsive and respectful of this customer (and I know of many others too) and his questions.
Anyhow, I got way off topic here. To get back to the initial post, I do wonder if more online examples or demo patches would help increase sales of a product. Or whether stores like Guitar Centre (in the US) or Long & McQuade (Canada) could have computer set-ups with the latest libs so that the potential buyer could really put them through their paces. The more we see the decline of hardware, the more other mechanisms for software driven libs/apps should be developed to allow the paying customer that much more information before they plunk down their cash.
Cheers (off to compose a string quartet).