Picking the right name for your indie game
How 439 people changed my mind twice and helped me confidently choose a name
Truly bad naming aside—it’s one thing to come up with some fine names, quite another to pick the best one. What will resonate with your audience? Will they read the same things into it as you? I faced this naming anxiety with my upcoming game and figured a way to make the decision confidently in one day with a $100 budget. I didn’t end up with the name I liked best—I ended up with a better one.
If you are facing the same challenge, let me share with you how it did it.
Step 1: Create a list of good names
A game’s name, along with its app icon and maybe a screenshot, is all you get to convince someone browsing an app store to take a closer look at your game.
- Keep it short: you don’t want to see your name truncated in the app stores.
- Keep it simple: easy to spell and pronounce, preferably common words.
- Make it enticing: it’s entertainment, after all.
- Make it unique: You want to be memorable, and no trademark lawsuits.
My working title was HUH?, which I thought was quirky and subtly descriptive. I came up with three contesting names:
- Gist: It’s about getting the gist of a level. Short, distinct and easy to spell.
- Intention: Similar to gist, but a more common word. Maybe a bit plain.
- Secret Signs: It’s about discovering a level’s secret meaning. Alliteration, nice rhythm, simple words, but a bit corny.
Step 2: Getting lots of feedback, cheaply
If you’re an established game company, you already have a large audience to ask for help. If you don’t, asking family and friends may be tempting, but probably won’t get you the unfiltered feedback you’re after.
Sites like UsabilityHub will recruit people to test your designs. Starting at $1 per response, getting a few hundred response is going to be expensive on an indie developer’s budget.
This is where I turned to Amazon’s Mechanical Turk (AMT). This rather awkwardly named service lets you define a simple task and offers a large audience to run it by. As is typical for Amazon, they take a low commission and aim for volume. I could get five times more testers while paying each one double of what they would make at UsabilityHub. With a $100 budget, I could get nearly 500 testers!
Beware: AMT is pretty bare-bones. You’ll have to build the survey and do the data analysis yourself. On the upside, you have full control! In the next section explains how I did all of that in a day.
Step 3: A simple and effective survey
I created my survey using Google Forms and tried to emulate the feeling of being in the iOS App Store:
Note how the question is phrased: I am not asking which name you like, but which game you would be most interested to play. After all, we want to know which name will make people take action, so try to place them in that moment. Make sure you enable the “shuffle option order” option. This eliminates people’s bias towards the first and last answer.
Next, I added 3 short questions.
What made you choose this name? We’d like to understand why people chose one name over the others. If lucky, we discover qualities or weaknesses we aren’t aware of yet. Be aware that people will often look for arguments to support a mostly subconscious decision process, so critically review the responses.
How often do you play puzzles games? Audience filters in AMT are expensive and broad. This question will show me if and how voting is affected by people’s affinity with puzzle games.
How often have you paid for mobile games? I’m considering putting a price on (part of) the game. This question will show me if and how voting is affected by people’s willingness to pay for a game.
That’s it! Let’s see what happened when I ran the test.
Step 4: A resounding result
AMT may not have the easiest setup or documentation (a topic for another article), but once it runs, it’s blazingly fast. I got all 220 responses in 15 minutes!
So what were the results?
Secret Signs received almost 50% more votes than the second choice. My working title and personal favourite HUH? didn’t even get half as many votes—damn!
But were the people voting for Secret Signs avid puzzlers? I used their frequency of play as a multiplier (from 0 for ‘never’ to 3 for ‘often’) and saw an even stronger outcome for Secret Signs:
A breakdown per name shows the neat correlation for Secret Signs between name popularity and people’s puzzle habits (granted, with some samples sizes too small for any statistical significance):
Lastly, I looked at how people’s willingness to pay affected their choice. It seemed to matter little:
If you’re curious to see how I turned my raw data into these charts, free to check out my Google Sheet!
Step 5: Ideating on the written feedback
Respondents were also asked to motivate their choice. As I mentioned, not all written feedback is helpful or reliable, but stay critical and you might learn new, like I did.
Looking through people’s writing, a trend emerged: Secret Signs elicited a sense of mystery. I thought it was fair to assume this was the main reason this name had won the first round.
“[Secret Signs] gives the puzzle an air of mystery or surprise”
“The idea of secrets is exciting.”
Inspired by this feedback and with help from Carli Hyland, I came up with another “mysterious” name: Object Unknown. I thought it sounded less corny and more distinct, soI was keen to pick it over Secret Signs. Controlling my impatience, decided to put it to the test once more...
Step 6: Challenged by majority, again
I replicated the first test with just the two names and got another 220 responses in under 15 minutes. The internet is truly amazing…
The result: my personal favourite lost again, receiving less than 1/3rd of the votes. Secret Signs was victorious once again:
And again, we see the same trend: more avid puzzlers preferred Secret Signs more:
I wondered what made people vote so strongly against Object Unknown. It only took a quick look at the written feedback to get a pretty good idea why:
“[Object Unknown] almost seems like an error — like the title didn’t load.”
“[Object Unknown] seems like it could be spam or something”
“Unknown sounds a bit fishy.”
Clearly, what I had perceived as a quirky kind of mysterious, was seen by many others as dodgy, broken or too vague. Secret Signs had won because it represented the unknown as an exciting discovery rather than a potential risk.
And so, the game was named Secret Signs and now lives on www.SecretSignsGame.com, currently as a demo with the first 10 levels!
In summary, my approach was:
- Create a few good names that meet basic requirements
- Create a survey with the necessary questions, and no more
- Cheaply test with a large audience using Amazon Mechanical Turk
- Optionally, use written feedback to iterate on more names
I hope you found some use or inspiration in my process. If you have any questions, feel free to reach out to me at firstname.lastname@example.org.
Epilogue: Yeh, but…
I suspect some people may have objections to parts of my method, so let me address a few preemptively :-)
“Yeh but, where’s your creative ownership when you let others decide on your game’s name?”
I came up with the names myself and was happy to us any of them. As a designer, I’m comfortable creating many options and using the input from others to inform my final decision.
“Yeh but, people never saw the game. What if one of the other names better describes your game?”
By only telling them it was a title for a puzzle game, I tried to figure out which name would work best in the context of app stores. Here, people will usually decide whether or not to look at your game by little more than an app icon and the name. And again, if you think a name does not describe your game well, don’t put it up for vote :)
“Yeh but, how do you know whether the respondents are the right audience for your game?”
Truthfully, I don’t. That said, if you’re aiming for general appeal within a genre like puzzle games, I suspect this kind of board sampling will give you a more accurate picture than, for example, asking a group of fellow game developers.
“Yeh but, are your results statistically significant?”
I don’t know. While I know how to A/B test, I don’t know how to test for statistical significance in this situation. What I do know is that I got a lot more data to base my decision on than I did beforehand, and I’m fairly confident I ended up with a better outcome because of it.
“Yeh but, it costed you money!”
I like free things just like anyone else. However, I am spending a ton of time on building this game, which is a big investment. I’m happy spending $100 to confidentially decide on a name and be done with it. If you have zero budget, you could try posting your survey on gaming forums or so. It may take a little longer and you have to be aware of other biases in your sample group, but it’s totally doable.