Here’s how he suggests developers make the jump from making good games to making great games:

What you need:

* A good knowledge base of ActionScript
* Either the ability to produce appealing art or the ability to hire someone to make it for you (Doesn’t have to be amazing)
* Time
* The desire to make a game where you can truly say “I want to play this” when you’re finished

If you don’t have any of the above making a great game isn’t going to happen for you.

Some things to point out:

* Genre does not matter, a good game can come from anywhere
* The key to a great game is execution, the original concept only has a slight impact
* Polish is extremely important, having bugs and loose ends can ruin an otherwise great game
* Your game should include a lot of content, games made in 1-2 days rarely become massive hits.

Alright, now to actually go through the steps of making a great game -

* Come up with a concept, copying previously successful base concepts is fine as long as you don’t leave too many similarities. Your concept should be more complex than “A Mouse Avoider”, make it very detailed. When doing this step I would suggest fully laying everything out (but always leave room for changes). You can also go the “Make it up as I go” approach, but this can often leave you uninspired when you don’t see where you’re headed. Make sure when the game is finished it has a lot of variety, nobody likes doing the same thing more than few times.
* Envision the end goal, you need to know that you’re going to make this game as good as possible.
* Start Coding
* Make/Collect the art, and add it into the game
* Finish a Prototype
* Get people to test your game, collect feedback and make changes accordingly
* Add more content (And by content I mean more unique features that vary the gameplay, not just ways to make the game longer)
* Have others beta test the game and look at their feedback critically, if you want your game to be great you need them to honestly tell you they enjoy playing your game just as much as other top hits. FGL’s FI system is very useful here.
* Always look at your game and say “How can I make it better”, “What can I add”, and “What do other games have that I don’t” and constantly make changes until your answers are “I can’t (be very honest, don’t get lazy and say you can’t because it would take too long)”, “Nothing”, and “Nothing” respectively.
* Balance the game, a good difficulty curve is crucial
* Polish the game making sure there are no loose ends and adding in little extras (These go a long way).
* Beta test again, leave no bugs or loose ends
* Get Sponsored, release the game
* Smile as you watch you MochiBot climb to millions of plays.

Now the above things will make your game well liked and spread well, but if you want a lot of plays you need to include factors that make the game replayable.

What makes a game replayable-

* Achievements
* Level Editor
* Randomness
* Lots of content W/ a save system
* Customization of something