Endowed Progress Effect and Game Quests

Imagine that two people, Kim and Carlos, notice that their cars are filthy and both go to the same car wash to make things right. With their wash they each receive a special card that lets them earn a free car wash if they get the card stamped enough times during future visits. Kim’s card says it requires 10 purchases for a free wash, but the perky girl at the counter gave her a head start with two free stamps. The card Carlos got doesn’t have any free starter stamps, but it only requires 8 future purchases instead of 10. So both Kim and Carlos are looking at the same number of purchases to score their complimentary car cleaning.

Who do you think is more likely to come back enough times to fill up his or her card? Kim or Carlos?

It turns out that it’s Kim, who got saddled with a card that required 10 total stamps, but who received enough free stamps to get her 20% of the way towards her goal. This is thanks to a phenomenon called “the endowed progress effect.” Basically, the idea is that when you give people just a feeling of advancement towards a distant goal, they’re more likely to try harder and try longer to reach that goal, even relative to people who have an equally easy goal but who got no sense of momentum off the bat.

Researchers Joseph Nunes and Xavier Dreze coined the term in a paper 1 where they did the car wash experiment described above. They found that 34% of people who got a 10-stamp card with 2 freebies ended up coming back enough to redeem the cards, compared to 19% of customers who started with an unstamped card requiring only 8 stamps. This despite the fact that both sets of customers only needed 8 stamps for a free wash. Nunes and Xavier also found that those endowed with the two free stamps tried to reach their goal faster by waiting less time between washes.

Buy 6 heartbeat sensors and win a chance to punch Bobby Kotick in the arm AS HARD AS YOU CAN.

Why? The researchers argue that the reason for the results is that by giving out free stamps, the merchant was framing the task (i.e., buying enough car washes to get a freebie) as one that has already been undertaken. There’s a substantial body of research that shows people are naturally motivated to complete tasks that they feel they’ve started and will want to remain consistent with previous intentions. 2 Other research has shown that the closer someone gets to completing a goal the more likely they are to increase their efforts towards closing that last little gap. 3 Apparently, giving people a couple of free holes on a punch card is enough to trigger both of these effects. 4

This has a few interesting possibilities for game design. Imagine, for example, that I’m playing through Fallout: New Vegas 5 and I get a quest to save 10 slaves from a nearby encampment. One way to deliver that quest to me would be to meet a NPC and have her say “Hey, there’s 10 slaves. Go free all 10.” And so I’d go off, and the quest would tick up “0 out of 10 slaves rescued, 1 out of 10 slaves rescued,” et cetera. Alternatively, if the game designer wanted to invoke the endowed progress effect, I could first receive the request upon opening the cell door for a pair of slaves on the outskirts of the encampment. One of the slaves could say “There were 12 of us altogether! Free the others!” and my progress would start off as “2 out of 12 slaves rescued” as the first two sprint off over the horizon. According to everything discussed above, I’d be much more motivated to complete this quest if it were presented this way.

Other examples aren’t hard to imagine. What if some NPC wanting 12 Goretusk livers in World of Warcraft gave me two to start with and raised the request to 14? What if, upon learning a new crafting skill that requires combining 5 widgets into one superwidget, the game gets me started with 1 widget and makes the recipe call for 6? What if, when I’m waiting impatiently in a multiplayer matchmaking lobby for Halo: Reach to find me 10 opponents, the game populates the first two slots with “Player Found!” after a couple of seconds even though it’s still looking? Would I be more likely to wait for the rest even if the search takes a long time? 6 Well, you get the idea. If you’ve got other examples, let’s hear them in the comment section.

Follow me on Twitter, Facebook, or RSS.


1. Nunes, J. & Dreze, X. (2006). The Endowed Progress Effect: How Artificial Advancement Increases Effort. Journal of Consumer Research. 32, 442-52.
2. e.g., Fox, S. & Hoffman, M. (2002). Escalation Behavior as a Specific Case of Goal-Directed Activity: A Persistence Paradigm. Basic and Applied Social Psychology, 24 (4), 273-285.
3. Hull, C. (1932). The Goal Gradient Hypothesis and Maze Learning. Psychological Review, 39, 25-43.
4. The more astute among you may guess that it has something to do with loss aversion –maybe those with the two free stamps valued them more than shoppers in the other condition and didn’t want to lose them. That’s a pretty good thought, but it occurred to Nunes and Dreze too. Without going into too much detail here, they did a follow-up study to test that hypothesis and found that the value of the endowment didn’t really affect whether or not people persisted in earning the free prize.
5. Which I actually am at the moment!
6. Sorry, is that evil? Lying to your players may be a little evil. Your conscience may vary.

16 thoughts on “Endowed Progress Effect and Game Quests

  1. So what explains these two Incompletes on my transcript that I have no intention of ever finishing? Is that because it turns out I can get my degree without them? Or is it because I feel satisfied with what I learned in the courses even if I didn’t get all the work done?

  2. LotRO may be doing that with deeds. After killing a few of X, you get a deed (a bit like a quest, for those not playing LotRO) for killing N of them. AFAIK that last kill that triggers the deed is counted towards completion, but I might be wrong.

  3. What’s interesting is that some automatch approaches I’ve seen on the Xbox 360 do say ‘player found’ for the first few players, without telling you their names. I believe Modern Warfare 2 does this.

    I’ve always wondered why. Do they not want you to look up the gamercards of the first few players and decide they don’t like them and quite the session early on? It would make a lot more sense if they don’t actually exist yet, so the game can’t show their name.

  4. Or (being facetious) turn it on its head: you are given a clip of ammo which can hold 30 bullets, but only contains 26 since 4 have already been fired. You are now more likely to empty the clip sooner than if you’d been given a full 26 capacity clip… ;)

    (Yeah, I know it isn’t the same situation; I’m just playing silly buggers).

    You can make the proper endowed progress effect more evil in a few ways. E.g.:

    Chaining quests together. Quest 1: You need 5 widgets are given 1 and are told to find Fred. You find Fred. Start of Quest 2: Fred wants to you take X items to Cuthbert, but only has 2 with him – go find the rest. Getting to Cuthbert requires killing a few kobolds (you’d better make sure it really does require this). Quest 3: Cuthbert thanks you for the items, and wants you to kill 10 kobolds, inluding the two you already killed getting to him (you still have to kill 10 kobolds in total, but you’ve got started on the quest without realising). He gives you a widget in reward. You now have 2 widgets out of the 5 you need, despite not having done any widget farming…

    The other evil variant is to give people a random number of ‘free stamps’. The not-so-bad variant is when Cuthbert is guarded by 1-3 kobolds, so you still end up having to actually kill 10. The nastier variant is when Fred gives you 1-3 of item X randomly. And you get to watch people on forums convince themselves that pure randomness is in fact determined by something they’ve done previously, and will swear blind that giving Fred a kiss before handing the quest in means that you get 3 items because it has happened to them every time (and the people making up stuff just to annoy other players). Cue the flame wars…

    The filling of game slots with non-people to give a sense of progress is quite cunning. Maybe world of warcraft should consider something similar for e.g. battleground queues. Rather than “estimated wait time 20 minutes”, give a value for how many people have joined the group already and how many more you are waiting for. And make it up if the numbers are too low to be enticing ;)

  5. Pingback: Super.licio.us | Superlevel

  6. @unwesen Really? Cool, well good on them I suppose.

    @Joost: Halo does it, too. I always assumed that it was downloading their info still, but I think you can actually highlight them and hit “Show Gamer Card” so I’m not sure.

  7. @Phil Regarding the gun thing: Well, only if emptying the clip were your goal. :)

    Also, good examples of other uses, especially the chaining stuff. The only thing I think designers would need to worry about is being too transparent.

  8. Pingback: Starting Users Down The Task Completion Path « Thoughts

  9. Isn’t the same effect at play when you download a free demo of a game and can later upgrade to the full, paid version? The publisher gets you started on the path towards owning the game and once you’ve acquired some of the content, you’re more likely to want to go the full mile.

  10. I disagree that the 8 stamp and 10 stamp cases are identical.

    If the car wash gives me two stamps, I’m inclined to believe they may do this again. (Maybe they don’t, but I’m optimistic.) Couldn’t this explain the effect, without bringing in the endowed progress effect?

  11. Pingback: Simoleon Sense » Blog Archive » Weekly Roundup 108: A Linkfest For The Smartest People On The Web

  12. I’ve seen this effect in Guild Wars, too, sort of.

    Even though GW has the option of playing only with AI in your party, it can be quite amusing to watch the LFG channel.

    Players with only 1/4, 1/6, or 1/8 players in their party are unlikely to get more members, even if there are enother 3, 5, or 7 players looking for groups for the same thing.

    The moment they have 2/4, etc though… suddenly their groups fill up really fast.

    I’d argue that this is also a certain degree of social laziness, not simply the endowed progress effect, but the latter probably does play a part, too.

    Never thought about that before! Neat! XD

  13. Hmm… The immediate question the study raises for me is whether or not they compensated for the perceived value of the card itself. If I’m cleaning out my wallet and have a punch card with no punches, it has no value and I’ll throw it away. I might still return to the car wash 10 times, but I no longer have the punch card. If it’s got punches, then it has a perceived value and will be retained.

    In terms of your quest examples, the different is probably negligible. I am, in fact, less likely to delete a quest from my quest log if I’ve made progress on it. So the effect applies either way.

  14. Every roleplaying game going back to tabletop games uses this dynamic. You always start with 50 gold pieces and a standard adventurers pack. Come to think of it, boardgames do to. Monopoly starts you with money and a game token.
    What’s new here is incentive to wait in queue.

  15. Pingback: 将赠与的进程效应整合到游戏任务中 - 爪游控

Leave a comment