Hiring an illustrator for your games cards can be expensive. I've had quotes anywhere between £10 per card and £300 per card. The complexity of the art style you're looking for, the experience and reputation of the artist and a host of other factors play into this figure. Given that a card game could easily have 54 cards requiring an image each that could be the difference between £540 and £16,200 over the course of a game. It's a big decision.
I'd like to talk about what a designer can do to make this part of the publication process easier, but first I'll offer an example from some of my past games to illustrate how hard this can be:
Escape the Nightmare raised less than £10,000. If we'd hired a top rate artist then the art costs alone would've been greater than the project raised. The art really needed to be on a budget for the project to be successful. Scandinavia and the World raised over £61,000 - we could've spent a significant amount on art and still had the project work out overall.
The trouble is this: For each game we had to determine who the artist was (and therefore our art spend per card) before the game was launched. A good project means showing a good game, which means showing some final art. Knowing what to spend on art is dependant upon knowing how well the game does because the cost is divided between all games rather than on a "per game" basis like manufacturing is. So you need to know how well the game will do before launch to make an optimum art decision.
Now in both of these examples we dodged the question on the publisher side. With Escape the Nightmare we used art from artists who'd usually charge closer to the top end of the scale, but we'd already paid for the commissions for another project and (with their permission) didn't need to use it again. With Scandinavia and the World we were partnered with a webcomic who got a share of the profits but provided all of the art. That's not always possible though, it hasn't been for more than half of our previous projects and probably won't be for our next one - so lets talk about what a designer can do.
Knowing that we weren't paying a cost per card on EtN and SatW my design brief was "Use as much art as you want" and the design of those games reflects it.
The situation for other games is different. The artist has to be chosen before how well the project is done is known, but the designer has control of the other side of the equation: How many pieces of art does this game need?
There are two philosophies that can work here: "Minimise art" and "Flexible art"
The first is simply to design the game in a way that requires the fewest possible pieces of art. If a game can say "Well there are five types of card and the piece of art on a card will depend on its type" then you can spend almost whatever you want on the art per card without meaningfully impacting the overall budget. This is often the simplest solution, but can make it harder for the game to deliver its theme and is something worth trying at a prototype stage (rather than prototyping with art everywhere knowing damn well it won't be there in the final thing) to see how testers find it.
The second and more complicated approach is to try to make the amount of art in a game flexible, so that the decision about the amount to spend on art can be made after the amount available in the art budget is known.
This is trickier from a design perspective, but the goal is to have cards that could have individual images or that could all have the same image. For example a game might have cards for "Fire bolt" "Fireball" and "Inferno" that could share an image or could have different images. If the project goes well then they get one image shared between all of the cards. If it goes exceptionally well then the extra budget can be used illustrating them individually.
I think this is fairly common, most likely as a result of designers coming up with games they'll pitch to several publishers who have different approaches to art. I don't want to name a game here because while I'd mean it as "Here's something sensibly designed to make sure gamers get the most out of their work" some idiot will take it as "I accuse this designer of being cheap" - but I'm sure you can generate lots of examples from your own games collection. Off the top of my head I can think of a dozen examples of game where cards with titles that imply they could be drawn individually, but that share art in a way that works and feels consistent with the theme.
The opportunity for designers to modify games to suit art doesn't end at theming cards to permit art duplication if necessary. It's good for design to magnify every aspect of a game wherever possible.
For instance with the Genesis project 3DTotal is very keen on bringing in a very high quality artist. That means spending a lot of money on art and a design that does the best to really show off and integrate that art is going to be important. There are a few games out there now using tarot sized cards rather than the traditional 63x88mm cards. That's all well and good - but a game has to be designed for that from the ground up!
The physicality of a card changes how they are used. A game with physically larger cards needs to minimise activities like shuffling that are harder with more cumbersome cards. It's also important to consider the amount of space a game needs on the table to make it playable in the environments you'd like to see it played. On the other hand it also presents opportunities - you can make more assumptions about what a player can see on their opponents cards from across the table for instance.
The point that I'm driving at with this post is that there are a lot of things that can be thought of as "The publishers problem" that are made easier or harder by the choices a designer makes in building their game. In a good game the design of the game itself and its physical nature and presentation are intertwined. It's worth being conscious of the pressures facing your publisher and of their strengths and limitations so that you can make the most out of working with them.