Unpopular opinion:

It’s not that hard to figure out what a client wants... most ideas are built on things that already exist; there’s a reason people say “it’s like Uber but for X”... ok... build them Uber for X... they told you what they want. Even if they don’t straight up tell you that, there’s still some piece of existing software that is doing something comparable to what they’re asking. You just have to understand how to implement it or build on top of it.

However if you do actually find yourself in unexplored territory, glhf cowboy!

It can be easy, and about quarter of the time it is. Some generic site, some specific utility, some artwork, or whatever product they need/want. Few changes requests, decent pay, good referrals. Treasure these clients.

It can be bloody impossible, too, because sometimes the client doesn’t freaking know what they want and will blame you for it, and ofc hate all of your attempts. Fire these clients. (Don’t worry about bad word of mouth; most people who know them already know what kind of person they are.)

Most of the time, it’s something pretty generic, but with endless scope creep and endless design changes because the client “will know it when [they] see it.” (Spoiler: they won’t.). Make these clients happy for awhile, but counter their endless requests with limited free changes. Be professional with them. The more professional you are, the better your experience will be.

i agree.

but in my experience the client needs advice. if he requests feature x, what does this even mean and what are possible drawbacks.
this is especially the case for not so tech savvy customers.

I value that my customers understand what they are asking for and this process is more or less hard depending on the person i am speaking to. they do not need to understand specific programmer stuff (that is what i am for) but they should get the bigger picture.