1 – Plan before you act
We are humans (crazy, right?), therefore we tend to work on things that we like, and procrastinate on the things that we don’t like. That’s one of the reasons many game projects fail. We end up developing games that WE like, instead of making games that others (the ones that will pay your bills) would like.
You can fairly ask me how would you predict what others might like? Think about how long you (and your team) can FINISH a realistic project. Then, go after business intelligence and trend analysis to help you make decisions about platform, genre, art, UX, game design etc.
Where you can find it? Google. Gamasutra. Google Trends. Statista. This kind of free tool, per example: https://games-stats.com/steam/
2 – Think your schedule from the end
Most of us, when starting a plan or a project schedule, go for a logical approach: start from the beginning. After two decades helping make games and publish them, I would advise you to do the exact opposite – start from the end.
Get a full understanding of your development cycle, taking into consideration your resources and budget (if any). Then, set up a fictional but realistic launch date. Stick to that, and move backwards. How many months before you should start creating buzz about the game? How long before the launch is the game ready for launch? Where the QA windows will be? Keep answering those questions until you get to the actual date you are in.
I can tell you right now that things will not fit immediately. Some tasks can be done simultaneously, but anyway you will need to adapt – and postpone your fictional launch. It is a very interesting exercise that I recommend to all my clients.
3 – Don’t be your own judge
Another very common behavior from developers, especially on the indie side, is to honestly believe they are doing the best game ever. I do agree to some extent. It’s your art, your dream, your “thing” – and it’s fine – but when people decide that because their idea sound amazing to their own ears, some sort of “blindness” happen, and this can definitively put an end on whatever chance the project had (even if was actually brilliant).
My advice here is: get A LOT of feedback, from the very start of the project, especially from people you don’t know (e.g. they will not say the game is great just to avoid hurting your feelings). Itch.io, Gamejolt, Reddit and several other channels are great for that, but make sure to have your own play test going on as well.
4 – Finish stuff
Make sure you have your capabilities and skills sorted out, to actually finish the project. Avoid opening concomitant projects unless those are paid jobs that will help the studio. There are examples of great games that were never launched and spent tons of money simply because the design and technical teams kept changing their minds all the time.
Commit. Make concessions. Great is the enemy of good, and good might be enough to generate the needed learnings to make something actually great.
5 – Don’t believe the hype
Oh, the hype. VR, Stadia, Geoloc Games, well, Anthem. See, be very careful to what is being “sold” by the media and major players in the market. Be even more careful before getting on the hype train. Remember, it is your job and your life at stake here.
Study trends and BI to make educated decisions, before jumping into something and crashing when the bubble pops. Or else, you may end up having your game do this to a player (it is an actual picture, taken by be, at a mall near my place, on one of those “VR Arenas” – I edited it because I don’t want to gross you out, lol).