#key performance indicators

Simplicity is actually super important when it comes to conveying information. Cognitive overload is definitely a thing, it causes frustration when there's simply too much for a player to pull in. UI and UX constantly have to toe the line between providing the information the player needs and overwhelming the player with too much information. Here's an example of what I mean - World of Warcraft's Raid UI is a deluge of information every frame. Only a player who has trained and practiced with it will be able to understand it at a glance, the sheer quantity of information (and all of its context) will absolutely overwhelm a new player.

AAA games especially tend to be among some of the most complex games because they have the broadest scope - we have the most content, the most features, and the most interlocking systems. This necessarily selects for simple and clean interfaces to lower the chances of overwhelming new players. The more we give the players to look at, the more cognitive load we're putting on them that may push them over the edge of frustration.

This isn't to say that games can't experiment with interesting UI - clearly they can. Persona 5 is well known for shaking up the UI world with their exceptionally stylish UI choices, many of which are being copied today with varying levels of effectiveness almost ten years later.

That said, the biggest factor holding back UI experimentation is that it isn't a major selling point. Very few games can sell on the stylishness or innovation of their UI - core gameplay and player retention trumps the UI as key performance indicators. Because UI is lower on the priority list, "good enough" is often the standard when it comes to allocating development resources. UI innovation really needs internal champion art directors to push for it and care about it in a world where "good enough" is good enough. That's a rare ask.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Short questions: Ask a Game Dev on BlueSky

Long questions: Ask a Game Dev on Tumblr

To build a brand new building you must check a number of components before starting construction. Working with a construction facility management consultants can certainly help alleviate your concerns about your project. Talk to these experts and learn a little more about brand new building construction and the dangers associated with them.

If you plan to build a new home or perhaps a large facility, you will go through a long process From the preparation process to the turnover of your home, you always have to observe the important things. You usually have to be by the side of your constructor to make sure everything is in the right position. When you forget to do this, you can waste your investment. You may ask yourself how you can protect your investment if you do not know anything about engineering. A construction facility management consultant can really help you with this process. They specialize in construction and have engineering degrees. They can come to your aid and help you explain the basics.

A construction consultant gives you a few solutions. Below are a few of them.

Project planning :

Such experts can guide you from the beginning of your project. They recommend designs and supplies to use so you know your new home or facility is of high quality. They help you create a budget so that you can spend based on the money you receive. This can be a problem for some people who do not get help and advice from the authorities. There is a tendency to spend extra on unnecessary additions to their benefits.

Facility management consultants also assist in overseeing current progress. Through this method, they help to identify potential hazards and present the method to them when they occur.

Forensic work :

Sadly, many buildings may not be as strong and durable because of certain components. If problems are not quickly identified, the entire building may collapse. Your consultant can really help point out the problem so that the developer can find a solution to it. This could be a weak spot on the floorboard or a problem with the foundation. They will try to find the root cause and plan the steps to prevent the problem from happening.

Security :

Generally speaking, people will buy games for the gameplay, the story, or the graphics and visuals. They'll buy games for the license or for the franchise (which usually means expected gameplay). They'll spend on a game for strong multiplayer and/or social features. .

Things people will not buy a game for usually entail the smaller elements of a game that are nice to have, but not sufficient to provide enough value for a player to spend full price on. These are features that engender good will to the players if these features are included, but is highly unlikely that many players will actively spend for. This includes features like:

UI

Voice acting

Music

Mini-games

Game/Video/Tuning Options

Photo Mode

Customizable/Mappable Controls

Accessibility Options

Cross Play

Cloud Saves

Subtitles

New Game+ or Replayability Features

Mod Support

Achievements

In-game collectibles

DLC

This is is by no means comprehensive - I'm sure you could think of additional features if you tried. This also doesn't apply to every game - mod support, for example, is a bigger deal in "maker" games where players make and distribute their own content.

If these features are present and "good enough", they will effectively be invisible to players. If they are wonderful, they might get special mention from happy players. If they are not present, players may complain a bit but their presence or lack thereof generally won't be the deciding factor in whether the spending decision is made.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Short questions: Ask a Game Dev on BlueSky

Long questions: Ask a Game Dev on Tumblr

And relatedly have you ever been on a project like the ME3MP you recently referenced where the team reportedly thought it was a nice little project but ended up blowing up beyond expectations? I seem to recall it made its yearly forecast in the first month or something.

I learned a long time ago that the majority of vibes, aura, whatever about how successful a project was going to be is all hogwash. I know what I like, but I can't tell which way a game will break. If I could do that consistently, I would be a much much wealthier person. I have a pretty good head for evaluating whether a game is good, but being good isn't anywhere near enough to guarantee success. All we can do is the best we can, and then see if the stars align and the public likes it.

That said, I have absolutely been on the other side of this - a couple of games I worked on and thought was solidly built but nothing super special actually ended up hitting it out of the park. I will say that the pay bonuses were really nice, but I was also experienced enough to know that I couldn't really do much more about it. It just as easily could have gone south instead through a thousand other factors outside my control.

The real takeaway from all this is that it's basically impossible to predict success. There's a huge number of factors outside our control when releasing a game - the state of the market, the interest of the players, the potential competitors in the market, and so on and so forth. We can't compare ourselves to a game that's lightning in a bottle like Arc Raiders or Slay the Spire 2. All we can do is hope that we hit our target numbers. Much of results are like a game of golf. Knowing how others do is helpful, but really we're only playing against ourselves. Other players cannot affect our score, they only serve as reference points. At the end of the day, only our own numbers matter to us.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Short questions: Ask a Game Dev on BlueSky

Long questions: Ask a Game Dev on Tumblr

Advertisers are canny when it comes to who they wish to advertise to. Different products appeal to different kinds of people - you're not going to see as many toy sales to adults as to children, for example. Since advertisers get to choose what kind of content they feel is good to advertise alongside, it will cause indirect pressure on any content to appeal to certain bands of advertisers. This will typically serve to rein in any kind of extremist content, since that is the kind of content that advertisers tend to shy away from. That's pretty normal for any kind of content - not just games.

Game devs will obviously consider trying to optimize for the advertising revenue, but it's an unsolved problem to work out. What's the tradeoff by adjusting the guardrails on the content in order to remain advertiser friendly? What are the key performance indicators and how do we hit them? Maximizing ad revenue would be to maximize engagement time with a sufficiently earning advertiser-friendly audience. How would one go about doing this? If there's to be an ad-driven gaming subscription service in the future, the winners would be the ones able to thrive within these kind of constraints.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Short questions: Ask a Game Dev on BlueSky

Long questions: Ask a Game Dev on Tumblr

Up until around 2014, preorder numbers were strongly correlated with financial success. They were used as the main metric for determining an upcoming game's chance of hitting its goals. Unfortunately, this caused a common feedback cycle problem with key performance indicators - the more you optimize for the metric itself, the less that metric continues to correlate to what you actually want to know. Publishers and retailers pushed various tactics to pressure customers to preorder, which did result in more preorders. However, the artificially-juiced preorder numbers didn't correlate as strongly with the success of the games. As lifestyle games grew into the biggest chunk of the market, preorder importance fell even more because it was the lifestyle game populations that were most strongly correlated with financial success.

The quality element you mention has nothing to do with preorders. Most games have two to four years of total development time before they ship. Preorders for those games often aren't available until a few months before the release date. By the time the preorders numbers actually come in, we're nearly finished with the development process. The timing doesn't line up at all, there is no way we can cheap out on quality in the last few months of development when the majority of that work is already complete.

As for whether I think players should preorder... I'll tell you what I do personally when it comes to purchasing games. I don't care much for preorder bonuses in general - they are almost always very small and won't affect the overall gameplay in any significant way. Preorder bonuses are generally limited to cosmetics at best, and the gameplay itself matters much more to me than having a particular outfit or weapon skin. There are some games I will preorder because I really like the game or franchise and I am concerned that the publisher has limited print runs of that game (Atlus, for example, is famous for having small print runs). Aside from that, I usually wait to see how the game does at launch before buying in. I don't feel much pressure to get in on day 1. Waiting a week or two won't kill me, and I've got plenty of other games to play and things to do in the meantime.

[Join us on Discord] and/or [Support us on Patreon]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Today I’ve got a war story from one of the developers on our [Discord server]. It’s an epic saga that encompasses years of development experience and investigation. I hope you like it as much as I did.

By now, it is fairly common knowledge that video games collect a tremendous amount of user data but, aside from financial reports, it is still fairly opaque about how those data are used. The following is a true story about how data made a tangible change in a live game.

When I started my job, I was asked to play through several hundred levels of our flagship title, a mobile puzzle game, as though I were playing for the first time. I wrote down thoughts and feedback about how I felt playing the game, while learning about how we made content. As I was playing, I got stuck on one specific level early on - Level X. It took me 44 tries to beat it. It was very precise, requiring the player to break specific pieces at specific locations in order to win. If I were a real user, this experience would have made me quit the game. I brought this up at the end of the week, and my lead revealed that "Level X is the best performing level in the game." No other level converted non-paying players to paying users as well as Level X. It singlehandedly caused the most purchases and generated the most revenue out of hundreds of levels. This shocked me. What I played caused only frustration - how could this level be our best? 

In addition to my normal work, I started asking about what data we gathered and why. Data informed how we created levels and designers were encouraged to understand it. I learned that data fell into two large buckets: Big-picture, business information to determine the health of the product and small-scale level-by-level data points. The first bucket includes concepts like revenue, retention (the percentage of people who keep playing at key time intervals - 7 or 28 days after their install date), conversion rate (percentage of free users who become spenders), daily active users, number of new installs. At the time, our second bucket full of small-scale metrics only included a few data points: Play Count, Win Rate, Quit Rate (% of plays that end before you run out of moves), and Monetization (revenue, in-game spend, power-up usage, and conversion rate). Level design primarily focused on win rate and level monetization.

I learned that Level X had a low win-rate of 4%, outside our normal difficulty target for the “hardest” levels at 5-10%. However, my lead was absolutely right - no other level came remotely close to Level X’s monetization metrics. Looking at the data that we had, everything shown indicated its low win rate was worth the revenue generated. However, something about looking at the levels around it puzzled me, because I could see far fewer play attempts on levels after Level X proportional to their difficulty. It seemed like we had an incomplete picture. Something about Level X didn’t sit right with me so I decided to investigate further. What followed was a multi-year process of revising how we gathered and analyzed data, and the most complicated spreadsheet I've ever worked on. 

The core of this effort focused on Churn Rate. Churn is defined as "the number of customers who leave your product within a given time frame." It took me, and another designer, and a product manager six months of working part-time to revise our metric tracking and validate it through live testing. We learned that Level X had a churn rate of an incredible 15%, double the rate of every other level in the game, and 4x the average of a level at its win-rate. When we presented our findings, the entire team sat in stunned silence for a few minutes - Level X defined an era of level design. Everyone strived to make a level that "performed as well as Level X." Were all of our premises about level creation misguided? What do we do now?

Come back next time for the conclusion to the tale - did they actually fix Level X? How could they do that?

[Click here for the conclusion]

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr

Today we conclude the guest tale that we began yesterday. You can read the first part by [clicking here], but here’s a quick summary - we’re at a mobile game studio whose flagship puzzle game has a legendary Level X that historically earned the most revenue of all of their levels... but was also so frustratingly difficult that 15% of all players who played it were so frustrated they quit and never returned. What to do?

The answer was “we try everything”. We didn't know how broad our changes had to be. Could we replace just the level itself? The levels around it? The entire early content path? How much time could we devote to this, while we still have a live game to run and update? Every designer on the project created multiple variations for Level X. We ran a sweepstakes for who could make the best-testing level. A splinter team started updating the rest of the first hundred levels. We ran dozens of A/B tests, where representative groups of players receive different content than a control group. Some were negative, some were inconclusive, some pointed us in a positive direction. We went through a full year of testing, and we'd finally come to a best-of test. It was our riskiest yet because we were changing the entire tutorial and difficulty pacing for the first 50 levels of the game - the entire new user experience.

For a month, the entire team was tense. The worst part about pushing test changes live is the waiting. Changing content requires getting new players to install and run your game, getting put into a cohort, and actively playing for weeks. Acting on early data could lead to huge repercussions. It influences revenue, retention, player count, player happiness - the health of your game as a whole. 

After that first month, we all breathed a collective sigh of relief. Our test cohort was leading in every one of the metrics. Level X was entirely different and, while it was still difficult and generated a third less spend, its churn rate was cut to 3%. The effect this had on levels after Level X was phenomenal. The block of the next 50 levels were seeing much larger amounts of players, increased spend, increased revenue. Our day 3 and 7 retention numbers were up by the highest we had ever seen in a single test. We pulled the metaphorical lever to release the content to every player.

Two years after I joined the team, Level X was replaced and the game was in a much healthier place. Our team's mindset changed to "If you build a game people keep playing, people will pay for it." Our design philosophy changed and we no longer made "pinch point" levels like the old Level X. However, no test is perfect, no content without flaws. In replacing levels, we made some better, some worse. We celebrate our wins, learn from our mistakes, and move on to our next test.

I want to thank my guest poster for sharing their war story with us. Our ability to gather feedback and refine our game design has evolved alongside the technology and online connectivity. Most dev studios now have analytics teams that track this kind of data and use it for both business and game design purposes. It’s valuable to see how players interact with the things we’ve designed because it can (eventually) lead to better player experiences overall.

[Join us on Discord] and/or [Support us on Patreon]

The FANTa Project is being rebooted. [What is the FANTa project?]

Got a burning question you want answered?

Short questions: Ask a Game Dev on Twitter

Long questions: Ask a Game Dev on Tumblr