If You Build It, They Might Come, Part Three: Encouragement

by Shannon Appelcline

September 19, 2002 - Welcome back! If you've been closely following this column for the last month or so, you'll know that I've been talking about player attention and retention. I started our in Trials, Triumphs & Trivialities #85, If You Build It, They Might Come, Part One: Attention when I discussed how you enticed players into your game with a great pitch that describes why it's cool. Then I followed up in Trials, Triumphs & Trivialities #88, If You Build It, They Might Come, Part Two: Discouragement by discussing how chaotic or annoying gameplay could lead your players right out of your game.

But, there's more. Just making sure that your game is clear and simple at start isn't enough to ensure that players will stay in your game. Or, rather, it doesn't allow you to retain the largest percentage of people who might be interested in your game. The method I want to talk about this week is encouragement — how you can increase the number of people who stay in a game by carefully offering them reasons to do so.

So let's get moving on that topic.

Conditioning and Rewards

A lot of the study of game design comes back, inevitably, to the study of psychology. Discussions of encouragement closely relate to the idea of operant conditioning, which I discussed at more length in another column in a galaxy far, far away: Thinking Virtually #34, The Psychology of Rewards.

In short: rewards are cool. People like rewards. If you offer rewards people will keep coming back in the hope of more rewards. In other words, rewards are a great way to offer encouragement to new players, which is exactly what you need to do to keep them coming back. I suggest the original Thinking Virtually article if you want to know more about some of the science we've learned underlying reward patterns. I'm only going to note one of these scientific points here, as being particularly relevant to the new player experience: Reward Consistently.

(On the other hand, you actually want to have some degree of randomness; the idea is just not to drop out a reward totally. But again see the original article for further discussion of that topic.)

If players take action A, then you reward them, then they take action A another hundred times and nothing happens, you're going to do the exact opposite of what you'd hoped. Players are going to get discouraged because they'll think they figured out a bit about how your game worked, when they got that first reward, then afterward will have realized they were totally mistaken. So, don't offer totally arbitrary or random rewards in an attempt to entice players into the game; rather make sure that players will be able to enjoy the rewards they'll regularly receive throughout the game starting in their first hour of gameplay.

Types of Rewards

I've sometimes heard this type of idea denigrated as a need for "instant gratification". Honestly, that's exactly what it is. If you don't offer some instant gratification in your game new players probably won't see any reason to stick around.

Unfortunately the idea of instant gratification runs straight in the face of a core ideal in many online games: socialization. I think the greatest reward in our own Castle Marrach is meeting cool people, getting involved with their lives, either IC or OOC, and really adding one (or five or ten) people to the list of "true friends" you have in this world. Even on games such as The Eternal City which feature more concrete rewards in the form of skills and money I suspect that ultimately people stay for the social bonds that they've formed.

The thing is, players are not going to form those social bonds in the first couple of hours of play. Thus, you need to be able to figure out other types of rewards that you can offer, that still make sense to your gameplay.

If you think carefully about your game design you'll be able to come up with ways to integrate early-game rewards into a social structure. You might offer a player an introduction to a prominent player as a reward; or tokens which allow them entrance into specific social gatherings; or autograph books which they use to collect character's John Hancocks.

However, it'll probably also be important to think about what non-social rewards might be a part of your game, and make sure those are easily available to new players at game start. Let your new players collect a number of non-unique items, or a single unique item. Let them increase their skills or abilities or statistics. If you're worried about score inflation there's no harm in making all starting players a little bit less experienced or able than you'd planned, then letting them get that initial boost early in the game.

In fact, it'll be a benefit because it'll let them see the cool things they can gain through game play.

Schedules of Rewards

I've been a bit vague thus far in talking about "early-game rewards" and I want to return to my core discussion of new player retention. It's my personal belief that you need to carefully manage player experience through the first month or so of gameplay, to ensure that they'll stick around as a regular player. To some extent that one month number is arbitrary, because it happens to be how long a Skotos trial account last. However I also think that a one-month time frame is a good measure for how long it takes to really get involved with a new hobby.

Are you more likely to keep going to the gym after you've done so for a week or a month? And then, is two months liable to make that much of a difference.

So, a month. It's actually one of five data points that I've laid out as important nodes for new player retention: a minute, an hour, a day, a week, and a month. Please do note, however, that they're all arbitrary, not dogmatic.

Rewards actually aren't very important for the one minute data point. You will lose a lot of players after they've been in your game for just a minute or two or five, but that'll have more to do with whether it seems to be their type or game or not and whether they're totally confused or not. See last week for that discussion.

So, let's climb up to one hour. By this time your player has explored a bit and hopefully understands whether your game is a good fit for him or not. But, he also needs some incentive to keep playing — that "instant gratification". By the one-hour point you should have offered your player the first chance to make his first improvement in his character.

• Perhaps he's been able to collect together some initial equipment and get equipped.
• If he's started fighting or otherwise using skills, he should on average have earned enough gains to increase that skill by an amount that he can truly see.
• If you're sticking with the strictly social type of reward your new player needs to have met some people and gotten a little bit involved in their lives.

Many games that I've seen center this one-hour goal around a tutorial. In Marrach, for example, the "tip" system will hopefully have led you to the seamstresses or tailors so that you get clothed and then have brought you down to the dining hall to meet people.

Grendel's Revenge now has a one-hour or so long tutorial in place that I hope will be very successful in retaining new players. It does a great job of not just introducing the systems but also rewarding the players. By walking through the entire tutorial you can get six or eight favors — enough to make that many small increases to your character — and some standard clothing that you actually get to select. You even get one totally unique item that's all your own. The whole time you can also participate in a "mentor" chat channel which allows you to make those first connections with other players.

The one day milestone will follow pretty shortly after the one hour milestone. An average player will play 60-90 hours a month, a bit more frequently than every other day. Thus you can probably count on a new player being in your game no more than 2-3 hours on the first day. You should probably thus offer a second reward within the second hour of play.

In all likelyhood the one-hour reward was related to character setup — equipment or something of the sort. Thus within the next hour you should make sure the player has that opportunity I already suggested to actually improve a skill or ability. It'll give him a real opportunity to see his character growing.

And hopefully that'll be enough to get him back after he logs out for the first time.

By the one week milestone, which is 15-20 hours in, you need to really make sure that a new player has had the opportunity to notably improve his character. Perhaps he's gained some minor social boon, like trial admittance to an organization. Perhaps he's not just improved a skilled, but improved it far enough to be able to access some new ability within the skill. Perhaps he's gotten a new skill altogether. If you have a level-based system he should have a second level by this point.

The point is, what your player's earned is pretty big, and it probably adds something totally new to the character.

Finally we come to the one month milestone which is the aforementioned 60-90 hours in. By this point, the next major level of achievement beyond that earned at the one week level should have been gained. Trial guild members become full members. Additional skills are gained and additional levels are earned. If there's an even higher level of achievement, this would be the time to reach it. For example, Grendel's Revenge allows wandering monsters to reincarnate at increasingly high levels of power. The one-month mark would be a good time to become a tier two monster. In The Eternal City it might be a good point to have achieved the full set of basic skills for a skill tree, all at usable levels. (Over in a strategic game like Galactic Emperor: Hegemony, within a month a player has finished a game, win or lose, and has had the opportunity to truly test himself against his opponents.)

One final note before I move on from the discussion of a reward schedule: consistency. It's quite possible that, in your game, people will start off at different levels of experience. For example in Grendel's Revenge it's possible to start a character with two or three abilities. If you offer this type of variation it's very important to still keep reward schedules the same for different characters. If a three-ability character takes a month longer to get a new ability than a two-ability character, then you're going to lose all your players who play three-ability characters because they won't have achieved sufficient rewards in their first month of play.

(The most correct answer here is to more carefully balance starting characters, but if that's not possible due to issues of complexity, you still need to balance the playing field for future skill gain.)

Remember the Fun!

This discussion of reward is all fine and dandy, but there's one other core idea you should remember: make sure your new players can do many of the things that existing players find the most fun

Does every one dote about the dueling system in your game? If so, you should make sure that newbies are encouraged to get involved with a duel early on. Are dark cults the core fun of your game? If so, make darned sure that new players aren't excluded. Is everything in your game centered around plots? Then figure out ways to incorporate walk-bys whenever you can.

In short: amuse them and they will come.

I have but one more topic to discuss regarding player retention: getting the help of your existing players. That'll be in 7. See you then.

Recent Discussions on Trials, Triumphs & Trivialities: