The first article discusses very general things about the contest, while the later parts are about the specific challenges the contestants face in R1 and later rounds. I hope it doesn't come across as too Sun Tzu. ;-)
Note: The article assumes that you're familiar with the general structure of the RPGSS contest and that you've read at least SKR's auto-reject threads. If you're new to the contest, please read Sean's consolidated advice thread first.
- Remember: it's a contest: It's useful to remember that in RPGSS, you're not designing for yourself or your gaming buddies. In R1, your item has to pass through two filters: the general public and the judges. Remember the go-ke, an excellent item and R1 exit poll favorite that surprisingly didn't even make the top 100, very probably because it included elements from a culture and game most voters are not that familiar with. Assess what people are likely to know. Have different kinds of people read your item, even non-gamers.
- Develop a strategy: Be consistent. Know what you're doing. Estimate what the other contestants are most likely to do. Determine your goals for the round. Decide what things you are going to do better than everyone else (everything, preferably). Look up the best entries from previous years and make a better entry than all of them combined. I know, it's easier said than done.
- Do not play safe: Tame is not Superstar. Take calculated risks but avoid dumb risks. In R3, I intentionally picked a monster that didn't get any recommendations from the judges. It was still a solid monster with good flavor, so it was perfectly workable. But don't go overboard. It's easy to mistake gonzo for creative.
- Teach yourself new things and skills: There is no excuse for submitting a map that looks like something a 7-year-old could have drawn in an hour. There's no excuse for not knowing about the thing about beards in Taldor just because you've never adventured in Taldor. Show you're capable of learning new things, and that you can adapt to new challenges.
- Test your design: If possible, playtest your item, monster or encounter. If not, at least have your friends read your entry and then ask them to describe it in their own words to a third person. If that third person still knows what's going on, you've probably succeeded.
- Don't forget what you learned before...: Many design principles you learn in R1 also apply to later rounds (but some don't). A lot of people who make the top 32 immediately forget to be awesome and creative in R2 and don't advance (it happened to me once). For example, if you create a monster that only re-uses other monsters' abilities, it's just as bad as a SIAC in R1. A monster with a seemingly random collection of abilities is just as bad as a SAK.
- ...But understand the differences between the different rounds: Monsters, villains, etc. are the GM's tools, so some R1 principles don't apply. For example, a monster that vomits is gross, but at least you don't have to wear it so it's probably ok. Also, RPGSS is a contest and R1 has a different voting system than the later rounds. For example, writing a 200-word item in R1 is ok. It is often actually smarter than using up your 300 words because many voters are more likely to read the shorter item first. But if you only use two thirds of your word count in the later rounds, you're probably not going to advance because you'll look lazy and you are wasting a chance to show what you can do.
- Use iterative design: Write (for example) four short item/monster/encounter concepts and talk about them with your friends. Ask them to pick one or two favorites and don't forget to ask why they liked the concepts. Develop one or two of the concepts further and have a second round of review.
- Store your best ideas for later use: Whether you succeed or fail, there will always be some leftover elements that didn't make it in your entry. Keep a google drive file where you store all your items, monsters and other material you didn't use, whether they're just one or two sentences long descriptions or fully developed entries. You might need them next year. You might "borrow" some of your own ideas in a later round. A cool item ability may turn into a monster ability or interesting terrain, for example.
- Be a gentleman/woman: Do not try to undermine other contestants' work. This includes insulting anonymous items. Instead of antagonizing other people, network, befriend them. They may be working in the same freelancing projects in the future, so you don't want to be rude because people have longer memories than you may think. Personality may also be a tie-breaker for some voters.