Whether you're a small-time indie developer or an AAA studio at the top of the food chain, you need to go through the playtesting phase. Try as we might, we'll never be able to truly view our creations in an objective manner - so if we want reliable feedback on a game, we need the opinion of folks who are not at all involved with it.
I've been having my game playtested for a while now, and with all the experience I've got and reading I've done on the matter, I feel ready to craft a small tutorial to help those who are just starting that step in their gamedev journey. Hopefully, you have a few friends or social media followers who'll be up to take a look at your game. You'll want to make the most of them, of course - this guide will help you make sure you do.
Let's start, shall we?
Setting Up Properly
As with most aspects of gamedev, the playtesting process does not start with playtest itself - there's a lot of things you need to keep in mind before getting people to play your game. Some pointers:
Watch testers play in real life, or at least use audio/screenshare
Many devs, when looking to get feedback for their game, will post it online or send it to their friends, and have them play it and get back to them with their thoughts on it. That makes sense. Players will probably be able to tell what changes you need to make to improve your game, right?
Wrong. Never do that.
The vast majority of people will be outright bad at describing their experience precisely, and most of them will refrain on being harsh on you if they have serious problems with your game. You need to watch how they interact with your game and what their raw reactions are - we'll see more on this on this later.
Finish whatever is in scope for the test
If you want them to test the mechanics, make sure they work 100% the way you've designed them. Don't be lazy and have them test something you know is buggy. Don't tell your players "Oh, that doesn't work right yet. It's supposed to do X, and Y" - go and fix that stuff before asking for their valuable time.
You need feedback on your intended gameplay. If you don't intend the character to, say, fail to double jump 50% of the time, it doesn't help you at all to watch how potential players react to that happening - not to mention it breaks flow and negatively affects their overall experience.
Make sure your tester has plenty of time to play
An important thing to measure when having someone play your game is how long does it take for them to lose interest. If someone has 15 minutes and you need 20 minutes of gameplay tested, leave it to another time. Their hurry will change their way they play, and you won't be able to see how long they'd want to play otherwise.
If your game already has some sort of replayability in it, or if you have more content than what's currently in scope for the test, allowing testers to play for more time than you need can be a great way to measure whether they really like playing the game - they will want to carry on playing if they they're enjoying themselves.
What Should You Be Looking At
Unless your game is being tested by another developer or someone with some sort of game design background, you should always take a playtester's suggestions with a grain of salt. When running a playtest, you are looking for opinions on how the game communicates with the player, and how does it feel on an unconscious level.
Focus on their actions, not on what they say.
On his 8 Good Practices of Writing, Neil Gaiman gives the following piece of advice:
Remember: when people tell you somethings wrong or doesn't work for them, they are almost always right. When they tell you exactly what they think is wrong and how to fix it, they are almost always wrong.
This is true for all sorts of art - an audience's gut a lot better at giving feedback than their brains. If a player reacts negatively to something in your game, or says it's "off" somehow, take it very seriously. But pay no mind to it when they say why is it wrong and how should you fix it - these are very hard to figure out, and doing so is your job as a developer, not theirs.
Notice how long the game takes to figure out
While the gaming is a learning experience (more about it on this excellent video from 3-Minute Game Design), you never want players to go through a long phase of figuring out how they're supposed to do stuff. If you ask a lot of them before they're invested, they will simply feel that it's not worth their time and effort and leave.
This is less likely to be the case with a playtester, since they'll probably make an effort to play and help you even if they're not having much fun - so you *need* to pay attention to how long it takes for them to get into action. Ideally, it should take no time at all - even if your game is complex, you should do your best to make gameplay start simple and intuitive, and slowly add each element in a way that players can grasp it quickly.
Obviously, it's no easy task predicting how easy will it be for most players to grasp a concept. The Illusion Of Transparency - a cognitive bias which makes it hard for us to put ourselves in the shoes of someone who knows nothing about what we're trying to communicate - will make you overestimate how easy it is for players to understand your game's rules and mechanics, so you'll need to fine-tune your gameplay a lot to have them learn as effortlessly as possible.
Whenever someone sits down to play your game, take note of which instructions or hints are learned quickly, and which ones go straight past them. I'm not sure how universal this experience is, but I often have a problem with HUD elements - I add important information there, and think it'll be really obvious to players, then 20 minutes into a playtest they get stuck - because they hadn't looked at it at all.
If you need help on how to direct players to what they need to understand, watch this Game Design Wit video on manipulating players' attention. It'll help a great deal.
Pay attention to the challenge/frustration balance
Unless you have a very casual game, players will often have trouble finishing each level. This is normal and desirable - but, in their frustration, they will often say things such as "This is too hard, you should make it easier". This is not the feedback you take into account: Instead, you should consider whether they remain interested or not, and whether they feel that a death was "cheap".
Pay attention to the flow of the game - players will often have trouble getting through parts that weren't designed to be real challenges, or breeze through parts that were supposed to be harder. If a challenge turns out to be easier or harder than it seemed in your mind, tweak them or switch the order so they get a proper difficulty curve.
What You Should Be Doing
Last but not least, there's a couple things you should be doing while your playtester playtests. These will help you get good feedback and keep track of it.
Take notes and do version control
Write down anything that doesn't match the exact experience you're hoping to produce with your game. If a tooltip goes unnoticed, write it down. If a player can't figure out how to interact with an object, write it down. If a challenge is too easy or too hard, if something's boring etc, write it down. Have a different list for each player, preferably with a small profile of them - so you can see what sort of player has problems with what - and a different list for each version of a game, so you can keep track of what was fixed and whether you have fixed it properly.
Stay back and don't interfere
People seldom have the developer of the game watching over their shoulder telling them what they're supposed to do or that they've missed a secret room. Allow them to play as if they were on their own, and resist the urge to say "it's a lot easier if you just do X!". Your game needs to communicate all important information by itself.
If there are secrets or clever ways to interact with stuff and they don't figure it out, that's fine! In fact, that's desired - it means the puzzle aspects of your games aren't obvious. In most case, each player will solve some of the intellectual/creative challenges on each run, but rarely get them all. Just keep an eye out for a challenge that no player ever solves, even after a few runs - unless you want it to be the secretest of secrets, you'll probably need to adjust and make it easier to figure out.
Getting feedback from players is the best way to fine-tune your game and make sure the gameplay experience is the one you had in mind - so don't skimp on playtesting. Hopefully, the tips in this guide will help you make the most out of your tests.
If you have any questions or want to add to the discussion, leave a comment! Feel free to hit me up or DM me on twitter (@WerneckXYZ) if you'd like to discuss playtesting or need some feedback for your game.
Thanks for reading!