The Art of Programming Conclusion: Just Look at Them Bones!
Our intrepid flatulaphonist has made it this far. He has decided whether he is acquiring audience members, or retaining the ones he's got. He has two rubrics in hand: machine guns and sniper rifles; rabbits and turtles. Now it's time to arm the troops for glorious combat in service to the motherland.
Context is everything
First off, the two extremes of the rubrics do not line up exactly with the two goals I have listed. For example, it could be that the machine gun style of programming is better suited to audience acquisition. Perhaps. But that does not mean that is its only use. Similarly, turtle pieces are perhaps better suited to audience retention, but not necessarily. Low risk pieces can also be an audience draw because they're probably pieces the layman has heard or can more easily identify with.
This is part of why programming is so difficult. Everything changes depending on what you're trying to accomplish. And we haven't even considered understanding the strengths of the conductor and the orchestra, yet. But lets ignore that and just think about how programs might be made with audience acquisition in mind. It is highly unlikely anybody has even heard of a flatulaphone, much less thought of attending a flatulaphone recital. We should name our heretofore anonymous flatulaphonist. His name shall be Fred.
Accuracy by Volume
We'll start with high variety programs. The idea here, if you will recall, is to present a large and diverse group of smaller pieces with the hope that something sticks. You don't particularly care that individual pieces resonate with the entire audience, just that everybody gets something they can walk away with going "Yeah, that Claire de Lune is such an awesome piece. I had no idea a flatulaphone could be so transparent and beautiful!" You also get the added benefit of discussion as audience members argue over which pieces were more worth their time. "What do you mean you like Claire de Lune?! That piece is such overplayed tripe. And it was in the Twilight movie! How can I possibly take it seriously? No, no, the Moonlight Sonata is where it's at."
You get bonus points if a fistfight starts, or you have to call security. This is in general, but nothing sells tickets faster than a good old fashioned audience riot.
Anyways. There's two directions I see that are valid. One is to take this type of program and fill it with turtles. The much maligned Andre Rieu does this all the time. He plays waltzes, polkas, and other generally popular and/or inoffensive pieces. The end result is a concert which appeals to just about everybody... at some point or another. But it's like cold calling. People don't remember the misses. They remember that part where it seemed like the medium communicated with them personally. That's where it's at.
The other direction I see is to take the opportunity to sneak in some rabbits. Higher risk pieces which may not resonate with many audience members, but you can get away with it because there are other sure hits which will cleanse the palate for those who found the rabbit's antics distasteful. There's probably a limit to what you can push on the audience. And yet consider: if you are specifically targeting newer audience members who are not yet acclimated to classical music culture, they will not know Schoenberg is almost universally maligned. Without previous expectations, they are more likely to listen with open ears. That doesn't mean they'll like it. But it does mean they'll probably give the risky piece a chance simply because "they don't know any better."
I personally like the second direction better because it will open up your programming options in the future. If the audience expects you to introduce pieces they've never heard before, but feel they can trust your taste in selecting them, you'll have a much easier time getting people to come back for more.
The Household Name and the Sniper Rifle
Lets say you want to attract audience members but still want to play big pieces. This is trickier, but not impossible. The best course of action here is probably turtles. If somebody is going to a concert featuring one or two pieces, they don't really want to spend their evening listening to somebody they've never heard of. Heavy hitters like Beethoven, Rachmaninoff, or Shostakovitch are all big draws.
The main issue here is, you can easily stagnate your repertoire if you're focused only on turtles. Over time, you become predictable, which isn't a bad thing in itself, but can lead to disinterest as you aren't presenting anything new and shiny to hold attention. It's all about the contrast. To that end, it's best not to focus entirely on one or two pieces. You might have one big symphony in the program, and then the rest of it is divide up machine gun style to keep things fresh.
A word of caution: it is tempting to view soloists as audience draws. While it's true the virtuosity brought by a concerto generally leads to good audience reaction, you cannot expect any random soloist to attract the attention of new audience members. Soloists draw sales because people can identify with them, because there's a face and a story that is familiar. Rostropovich wasn't world famous just because he was the best cellist of all time. He was also a political mover and shaker, had a personality, had gravitas. Soloists today usually just have a list of awards. That's not a story. That's a resume. If you're trying to attract new people with a soloist, you'll have to do better than that.
The Rabbit in the Rain
It may seem like the rabbit gets the short end of the stick, here. It's difficult. Risk is always difficult. But risk must be taken to keep the art form alive. The best way to turn a risk into not as risky is to make it a regular feature. If people come to expect modern and contemporary pieces, then they will come for that. There are, however, some difficulties. Certain repertoire is so difficult for the audience that it will simply never get performed, even if those pieces are masterworks. Here is where a well designed educatory program can help the most. That's another discussion, though.
Audience attraction will depend greatly upon a balance of pieces they are familiar with and pieces that are new to them. You want pieces that are riskier in your programs because you can't become staid and predictable. That's where the audience retention part comes in. If people always know what to expect, they'll lose interest. At the same time, you can't be doing random things all the time. Familiarity and novelty are your two main tools, within whatever program structure you choose.
Finally, a word about pacing. You want to start strong and end strong. Stuff in the middle can be whatever, but should still have shape. Programming is a lot like composition. You're taking the audience on a ride with ups, downs, curves, and loops. Slow pieces and fast pieces and middling pieces should all be carefully placed to help keep the audience's interest. Modality is also a big issue, because it lets you create progression through the program. What I'm trying to say is, if you've got a piece in Brown Note Minor, you should save it for the big climax at the end. It'll make for a performance no one will ever forget.