游戏化应用的兴衰:心理学、科技与AI经济的演变 TechButMakeItReal 2026-03-05

未完成的诱惑:游戏化心理的起源

当一个人年仅九岁,却已是晚上十一点,在一个本应乖乖睡觉的学夜,藏在被子底下玩了两个小时的《俄罗斯方块》。这个游戏没有角色,没有情节,完成时也没有奖励。然而,每一次成功地将一块方块放入槽中,使其消失,都会在大脑中激起一阵强烈的满足感。这种难以言喻的愉悦感驱使着玩家持续下去,即使早已失去享受的乐趣,仅仅是为了等待那一块必将到来的、必须契合空隙的特定形状。1984年,那位发明《俄罗斯方块》的苏联工程师,或许并未意识到自己设计出了一个在心理学上最具力量的系统之一。数十年后,研究者们开始深入剖析《俄罗斯方块》,试图理解为何人类大脑如此难以割舍那些未完成的模式。如今,历经四十年,科技公司们则将这种本能,这种对未完成事物所带来的不适感与张力,巧妙地融入了我们手机上的每一个应用。那些为了错过一天而惩罚你的“连胜记录”(Streaks),那个提醒你西班牙语课已逾期的猫头鹰,以及那些奖励你进行“并非真实”之事所获得的“徽章”(Badges)。而这一切之所以奏效,是因为它们确实有效。今天,我们将深入探讨游戏化的崛起,人们为何对此感到厌倦,游戏化应用背后的商业逻辑,以及它在AI经济时代将被何物所取代。让我们一同深入探索。

Original English

You're 9 years old and it's 11:00 p.m.

on a school night. You're supposed to be

asleep, but under the covers, you've

been playing Tetris for 2 hours. There's

no characters. There's no story line.

Nobody's giving you a prize when you're

done. But every time you drop that piece

into a slot and four lines vanish,

something fires in your brain that makes

it feel incredibly satisfying. So, you

keep going. You're not even enjoying it

anymore. You're just waiting for that

one shape of a piece that has to fit

into the gap. The Soviet engineer who

invented Tetris in 1984

didn't know he was designing one of the

most psychologically powerful systems

ever built. But decades later,

researchers would study Tetris to

understand why the human brain can't let

go of unfinished patterns. And 40 years

later, tech companies took that same

instinct. The discomfort of almost the

tension of unfinished and wired it into

every app on your phone. Streaks that

punish you for missing a day. An owl

reminding you that your Spanish lesson

is overdue. Badges that reward you for

things that are not real. And guess

what? They do it because it works. Today

is about the rise of gamification. why

people are becoming sick of it, the

business of gamified apps, and what will

replace it in the AI economy. Let's dive

in.

历史的脚步:游戏化原则的萌芽

或许出乎意料,游戏化(Gamification)这一概念的出现,早于其正式命名已有百年历史。其核心原则根植于基础的人类心理学强化原则(Reinforcement Principles: 通过奖励驱动特定行为的心理学理论)。若以代币(Tokens)等形式给予人们某种奖励,他们往往会倾向于执行期望中的行为。

历史上,最早的游戏化例子之一便是1896年的“绿色邮票”(Green Stamps)。S&H绿色邮票是一个传奇性的忠诚度计划:顾客通过日常购物获得邮票,这些邮票可以用来兑换S&H目录中的商品。到了20世纪60年代,S&H发行的邮票数量竟然是美国邮政服务的三倍。这些邮票并非毫无价值,因为集满一整本邮票足以换购真实的商品。但更奇特的是,在累积到有意义的价值之前,这些代币本身就已成为情感上的重要之物。人们会根据能获得的邮票数量来选择超市,家庭甚至会专门辟出厨房抽屉来存放这些集邮册。这种行为并非由投资回报所驱动,而是源于对进步的心理(Psychology of Progress)和最终能够兑换一叠积分的期待。

近一个世纪后,美国航空推出了“Advantage”计划,成为美国首个常旅客计划(Frequent Flyer Program)。该计划用通过预订系统追踪的“里程”(Miles)取代了纸质优惠券。次年,Holiday Inn紧随其后,推出了首个酒店忠诚度计划。短短五年内,积分与里程计划的爆发式增长席卷了航空公司、酒店和租车行业。这些计划证明了,人们不一定需要实物奖励才能注册或购买。即使是那些不可触摸、没有名义价值的非物质化积分(Non-tangible Points),也能驱动用户的忠诚度。这一行为模式随后迁移到了科技领域,最终导致游戏化泛滥的“流行病”,正如你可能已经猜到的,便是在线游戏的崛起,以及排行榜(Leaderboards)、连胜记录(Streaks)、积分(Points)、社会地位(Social Status)的引入。这也正是我们的游戏化与软件故事的开端。

2005年,微软推出了Xbox成就(Xbox Achievements)和玩家分数(Gamer Score)。此举的意义在于,平台上每一款游戏都为其贡献了一个单一的分数,这个分数对所有玩家可见,并能在多款游戏中通用。这一创举将奖励的概念提升到了“元奖励”(Meta Rewards)的层面,它超越了任何单个游戏的范畴。这一创新也改变了游戏行为,因为它激励用户去探索更多内容、完成支线任务、甚至在更难的难度下重玩游戏。可以说,Xbox成就系统是第一个大规模市场验证,即零货币价值的虚拟徽章能够重塑数百万人的行为。进入21世纪10年代,游戏化开始突破游戏的界限,迁移到了社交媒体领域。

Original English

This may come as a surprise, but

gamification existed for a century

before anyone came up with that name.

The core principle is taken from basic

human psychology and reinforcement

principles. If you reward people with

something, with tokens, for example,

they tend to perform desired behavior.

One of the oldest examples of

gamification that history remembers till

this day is the green stamps in 1896.

SNH green stamps were a legendary

loyalty program. Customers would get

stamps for everyday purchases that could

be redeemed for products from an SNH

catalog. By the 1960s, SNH were issuing

three times as many stamps as the US

Postal Service. The stamps were not

worthless because a full book of stamps

could buy you real merchandise. The

peculiar part though is that the tokens

themselves became emotionally valuable

long before they added up to anything

meaningful. People chose supermarkets

based on the number of stamps they would

get and families devoted entire kitchen

drawers to those booklets. The behavior

wasn't really driven by the return on

investment. It was driven by the

psychology of progress and the

anticipation of eventually cashing in a

stack of points. Almost a century later,

American Airlines launched a advantage,

which became the first frequent flyer

program in the United States. The

program replaced paper coupons with

miles that were tracked through a

reservation system. The next year,

Holiday End followed with the first

hotel loyalty program. Within five

years, a boom in points and miles

programs took airlines, hotels, and car

rentals by storm. These programs proved

that people don't necessarily need a

physical reward to sign up or to buy.

The motivation can occur with

non-angable points, points that can be

touched or don't have any nominal value,

but still drive loyalty. This behavior

migrated to tech and the last step

towards the epidemic of gamification

became as you might have guessed the

rise of online gaming and the

introduction of leaderboards, streaks,

points, social status and this is where

our story of gamification and software

begins. Microsoft came up with Xbox

achievements and gamer score in 2005.

When they did that, every game on the

platform contributed to a single score

that would work across multiple games,

visible to all players. This single

action elevated the concept of rewards

to meta rewards that would exist outside

any individual game. This introduction

also altered gaming behavior because it

drove users to explore more, do side

quests, replay games on harder

difficulties. Xbox achievements were

arguably the first mass market proof

that imaginary badges with zero monetary

value could reshape the behavior of

millions of people. In 2010s,

gamification started leaving the borders

of gaming and migrated into social

media. Now, this might tell you how old

I am, but I remember the days of

Foursquare. For the younger demographic

on the channel, Foursquare was a social

media app where you could earn points

for checking into certain places, gyms,

restaurants, cafes, coffee shops, and

you would be labeled according to the

places you visit. For example, a gym rat

or a coffee coner. Then came the

legendary game called Farmville. And

then after Farmville, the Silicon Valley

began showing interest in gamification

startups. The gamification market grew

200% in 2 years between 2010 and 12.

Deote cited gamification as one of its

top 10 technology trends of 2012. Then

came the first university level

gamification course at Wharton. Numerous

online courses across various platforms.

And this is how in the span of 24 months

gamification went from a strange word to

having a dedicated spot in Gartner's

hype cycle. It became a $40 million

startup category and the default

prescription by consulting firms. If

your product was struggling, add points,

add streaks, and add leaderboards. Now,

as a product manager, like the vast

majority of product managers in the

field, I had a time when I went through

a classic set of product management

books. One of them being Hooked, how to

build habit forming products. Now, this

book is fairly controversial, always has

been. The core of it is based on the

concept of a four-step loop. Trigger,

action, variable reward, and investment.

The loop is drawn from gambling

psychology. Variable rewards create

dopamine-driven anticipation. And the

idea is that you apply this loop to

consumer software. There are a lot of

different opinions around the ethics of

this principle. Some people being more

pragmatic and some people criticizing

it. As I was doing research for this

video, I found a comment on Reddit that

sums up the opinion of the latter group

pretty well. My biggest problem with the

book is its basic premise that hooking

people, that is making them compulsive

users of your technology product, is

something worth doing. Now, by no means

am I moralizing. You guys know I try my

best to be as neutral as possible. To me

personally, the shocking aspect around

the popularity of this book is the

realization that billions of people get

a Pavlovian level feedback from a fairly

basic concept and the fact that smart

product teams who work at technology

companies figured out how to ring a

bell. This is the reason why I consider

gamification phenomenon quite

fascinating because of how basic it is

and because of how people figured out

what to do with it.

伦理的审视:习惯养成与争议

“我最大的问题在于这本书的基本前提,即‘钩住’人们——也就是说,让人们成为你技术产品的强迫性用户——是值得去做的事情。” 这段来自Reddit的评论,恰如其分地概括了一部分人的批评观点。需要强调的是,我绝非在此进行道德说教;你们知道我一直致力于保持中立。对我个人而言,这本书之所以广受欢迎,令人震惊之处在于,数十亿人竟然能够从一个相当基础的概念中获得巴甫洛夫式反馈(Pavlovian Level Feedback: 通过条件反射产生的反应),并且技术公司的顶尖产品团队竟然找到了“摇铃”的方法。这正是我认为游戏化现象引人入胜的原因:因为它如此基础,又因为人们挖掘出了它的潜力并加以运用。

到了2014年,游戏化已经从一个游戏术语,演变成面向企业的全套功能集合。Gartner(高德纳咨询公司)在报告中指出,游戏化被用于企业培训,将获得认证的时间缩短了50%。Salesforce(销售易)通过成就追踪来游戏化其销售流程。LinkedIn(领英)则推出了个人资料完成度和课程徽章系统。然而,这场狂热的顶峰,无疑是一个你我都在耳熟能详的公司所代表的——这个公司体现了游戏化应用的单一最重要案例研究:Duoingo(多邻国)。

作为毋庸置疑的明星,Duoingo将游戏化机制推向了新的高度,其中蕴含了许多值得注意的细微之处。许多其他学习应用试图在各种方面复制Duoingo的成功,甚至在营销中提及Duoingo的名字,例如“Duoingo,但用于历史”、“Duoingo,但用于数学”、“Duoingo,但用于科学”。这里的重点在于,Duoingo开发的模式,其核心并非关于学习语言本身,而是一个“恰好能教授语言的游戏”。

这里还有另一条有趣的评论:“我坚持了555天学习中文后退出了Duoingo。我什么都没学会。” 这句话精辟地指出了问题的关键:“游戏化就是当你最终为了分数而奔波,而不是为了知识,并且那只小猫头鹰还会因此责备你。”诚然,我可以学会说“你好”和“你的电话号码是什么?”,但却无法进行一场基本的对话。确实如此。就我个人而言,我从未对Duoingo这类应用产生兴趣,因为它与人类语言习得的机制毫无关联。收集猫头鹰和维持连胜记录,与撰写和说一门语言的认知能力毫无关系。尽管如此,我必须承认,Duoingo是全球使用最广泛的语言学习应用。我敢肯定,在座的许多人会与我持不同意见。

目前,围绕游戏化疲劳(Gamification Fatigue)已形成一股普遍情绪。要理解为何我们会对此感到厌倦,让我们回顾一下它最初是如何以及为何发生的。有三个关键因素使得游戏化的兴起成为必然。

第一,智能手机。在智能手机出现之前,游戏机制需要一台游戏机或电脑。无论游戏引擎设计得多么精妙,它们都要求你必须身处电脑桌前。而智能手机,则使这种奖励行为(Reward Behavior)得以全天候可用。

第二,数据基础设施。现代应用能够衡量一切:会话时长、用户留存率(Retention)、流失率(Churn)、生命周期价值(Lifetime Value)。游戏化为产品团队和科技公司提供了可供操纵的具体杠杆,以影响这些关键指标。

第三,注意力经济(Attention Economy)。随着应用数量的激增,以及智能手机在我们手中日益根深蒂固,用户留存变得愈发困难。游戏化成为了维持用户注意力的最廉价方式,而无需研发需要大量研究和工程资源的新功能。

游戏化在游戏行业中之所以能够奏效,是因为在游戏中,努力与回报之间的关系非常透明且可预测。不像老虎机,玩家知道玩得越多,收集到的分数或金币就越多。你为什么要这些分数?比如,是为了升级角色。当你升级角色后会发生什么?你在游戏中的地位会提高。这在现实生活中意味着什么吗?不,毫无意义。但它在游戏中拥有意义。

现在,将其应用于健身应用。例如,你坚持了30天打卡一份膳食计划,累积了大量积分,可以解锁为期5天的会员版本,然后在月底获得一枚可以挂在成就墙上的徽章。心理学和神经科学数十年的研究已经证实了一个现象:在奖励的预期阶段,多巴胺的分泌最为旺盛。早在60年代,在这些心理机制拥有正式名称之前,“绿色邮票”就已证明,累积奖励、累积预期、收集幻象以及悬而未决的张力,改变了80%美国家庭的购买行为。此后,从常旅客计划到Duoingo的连胜记录,每一个游戏化系统,都是那一本“手册”的后代。

Original English

My biggest problem with the

book is its basic premise that hooking

people, that is making them compulsive

users of your technology product, is

something worth doing. Now, by no means

am I moralizing. You guys know I try my

best to be as neutral as possible. To me

personally, the shocking aspect around

the popularity of this book is the

realization that billions of people get

a Pavlovian level feedback from a fairly

basic concept and the fact that smart

product teams who work at technology

companies figured out how to ring a

bell. This is the reason why I consider

gamification phenomenon quite

fascinating because of how basic it is

and because of how people figured out

what to do with it. By 2014,

gamification had evolved from a gaming

term into a full-blown bucket of

features for enterprises. Deote used

badges to gamify executive training and

time to certification dropped 50%.

Salesforce gamified sales pipelines with

achievement tracking. LinkedIn

introduced profile completion and badges

for completing courses. But the pinnacle

of this mania is the company that

everybody has definitely heard of. The

company that embodies the single most

important case study for gamification

use. Dualingo.

The undeniable star, Dualingo, that took

gamification mechanics to the next

level, has a lot of interesting nuances

that are very much worth noting. Many

other learning apps try to repeat the

success of Dolingo in many ways,

including using Dolingo's name and their

marketing campaigns. Dualingo, but for

history. Dualingo but for math. Dualingo

but for science. Point is, Dolingo

developed a model that isn't about

learning a language. It's a game that

just happens to teach a language. Now,

here's another interesting comment that

I came across. I quit Dolingo after 555

days of learning Chinese. Didn't learn a

damn thing. Gamification is when you end

up farming points instead of knowledge

and the little owl bullies you about it.

I can say things like nihow and what's

your phone number? But I can't hold a

basic conversation. True. On a personal

level, I was never into apps like

Dualingo simply because it has nothing

to do with the mechanism of language

acquisition in humans. Collecting owls

and maintaining a streak has nothing to

do with the cognitive ability to write

and speak a language. Nevertheless, I

got to give it to Dolingo. They are the

most frequently used language app on the

planet. And I'm sure lots of you would

disagree with me. Now, there has been a

wave of general sentiment around fatigue

from gamification across various apps.

And to understand how we got to the

point when we're sick of it, let's go

back to how and why it happened in the

first place. There are three factors

that made the rise of gamification

inevitable. Number one, smartphones.

Before the rise of smartphones, game

mechanics required a console or a

computer. The games, no matter how well

engineered, required you to be

physically present at your desk.

Smartphones, on the other hand, made the

reward behavior available 24/7. Number

two, data infrastructure. Modern apps

can measure everything. session length,

retention, churn, lifetime value, and

gamification gave product teams and

technology companies specific levers to

pull against those metrics. And number

three, the attention economy. As apps

multiplied, as smartphones became more

ingrained into our hands, retention

became harder. Gamification was the

cheapest possible way to hold your

attention without building new expensive

features that would require lots of

research and engineering resources. The

reason why the reward behavior worked in

the gaming industry is because in a game

the relationship between effort and

reward is very transparent and

predictable. Unlike a slot machine, for

example, a gamer knows that the more

they play, the more points or coins they

will collect. What do you need the

points for? To upgrade a character, for

example. What's going to happen when you

upgrade a character? You get a better

status in the game. Does it mean

anything in real life? No, it doesn't.

But it has a meaning in the game. Now,

apply this to fitness apps. For example,

you stick to a meal plan for 30 days in

a row. You collect a bunch of points to

unlock premium version for 5 days, let's

say, and then you get a badge at the end

of the month that you can put up on a

wall of achievements. And the phenomenon

that's been confirmed across decades of

research in psychology and neuroscience

is that the largest dopamine surge

occurs during the anticipation of a

reward. Green stamps proved back in the

60s before any of these psychological

mechanisms had formal names that

stacking rewards, stacking anticipation,

collecting illusions and unresolved

tension changed the purchasing behavior

of 80% of American households. Every

gamification system ever since, from

frequent flyers to dual lingo streaks,

is a descendant of that standbook.

So why did it collapse? The

psychological tricks that sparked the

initial excitement became the very thing

that burned users out. Gamification that

was supposed to be a fun element that

would make the product less boring and

more organic to use and help you learn

things by playing became a commodity.

During the wave of fascination and wild

growth of gamified apps, Gartner made a

prediction in late 2012 that 80% of

gamified apps would fail by 2014. And

the reason for it was poor application

design. This is very similar to the AI

bubble that we're witnessing right now

and the video we made on why most AI

startups are bad businesses. Because you

can only get so far when pursuing trends

and not having a solid product. Remember

how I said that gamification was

inevitable? Smartphones, data, and

attention economy. These three

conditions that made it inevitable are

the very conditions that also made it

unsustainable. When the iPhone came out

in 2007, the App Store had about 500

apps. By 2015, Apple and Google each

hosted around 1.7 million, but the

number of apps people actually used was

around 27. And that number plateaued

even as time spent on phones kept

climbing. Which means that apps were

competing for 27 slots on a person's

iPhone. This environment made

gamification the cheapest available tool

for retaining users. Points, badges,

leaderboards, streaks cost next to

nothing to implement compared to

building genuine product

differentiation.

And the problem is it works in the short

term, but still it works. Introducing

gamification as a patch fix was rational

for businesses. Speaking from

experience, after building a career in

product management, sometimes you have

to find the cheapest possible way to fix

a problem. But this approach comes at a

cost. for gamification. It got bolted

onto apps as a retention hack rather

than building a business with a

defensible value that won't be wiped

away by hype cycles. And what's even

worse is that it disfigured the notion

of games because games are not about

points or coins or badges. These things

are as far from games as they are from

fitness or language learning. Streaks

are the most effective retention

mechanic in modern gamification. And

they also have a destructive second

order effect. They convert genuine

interest into anxiety. Even Dualingo

admits this. Long streaks motivate

through loss aversion rather than

positive anticipation. What this means

in playing English is that the streak

doesn't make users want to learn more.

It means that they're afraid to stop and

therefore they end up doing the absolute

minimum to simply preserve the streak.

But preserving the streak has nothing to

do with learning. The streak is there,

but you're learning essentially nothing.

The gamification of learning in mobile

apps has replaced the product's actual

value. Forsquare is the clearest

business case of gamification fatigue.

The app had 45 million users in 2013,

and all of that audience was built on

check-in badges and mayorship

competitions. But by 2013, the novelty

of checking in wore off, and the

gamification could not sustain the

interest because the value of the

product outside of gamification simply

wasn't there. So, how come Dolingo

didn't follow the same path?

Dolingo is a notable exception because

they invested over 600 experiments to

get gamification mechanics to perfection

and essentially outrun the trademill. I

can have whatever opinions I want about

Dolingo's learning mechanics, but I got

to give it to them. They managed to

build a stellar system for discovering

and testing behavioral signals faster

than any other competitor. In behavioral

economics and psychology, there is a

term called hedonic adaptation. And

Dolingo manages it beautifully. It

constantly refreshes visuals, rules,

rewards, streak freezes so that the

brain never fully habituates to the game

mechanic. They make their app a living

system. On top of it, they built a very

defensible technical infrastructure.

They tracked micro behaviors across

hundreds of millions of users. Most

gabified products back in 2010s never

invested at this depth. And even now,

compared to today's market, Dolingo's

product culture is rare. Yes, they built

their business on points and badges, but

they also built the data science and

experimentation plumbing to understand

why it worked or failed. Now, from the

language learning app standpoint, and

some of you may disagree, but I

personally do not see them as a product

that has anything to do with

linguistics, but that is a subjective

opinion. Nevertheless, Dolingo is a

notable exception. Most companies don't

have the resources or the culture to

sustain that level of experimentation,

which is why the Gartner 80% failure

rate proved accurate for the broader

market, but not for them. Now, where is

gamification headed in the AI economy?

Despite all odds, the gamification

market is projected to grow from $20

billion to $190 billion by 2034. But the

term called gamification eight years

from now will bear no resemblance to the

points and badges system that we see

today. AI will personalize the hell out

of it to the point when there is going

to be a built-in behavioral layer inside

each app. So what exactly will happen?

Number one, personalization is going to

be on the user level. Social media,

gaming, e-commerce will be reading

behavioral signals and know exactly what

motivates each individual user. When

they're about to disengage, how to

change the experience on the spot, what

to put in your notifications, how to

adapt lesson difficulty, how to change

your training plan depending on how

you're feeling today. What AI will do is

mimic the system that Dolingo pioneered.

AI will turn lots of products into

living systems that will constantly

change and adapt. Number two, listen

closely. The definition of loyalty will

be reframed. Latest research and

predictions around transactional

loyalty. Transactional loyalty being

discounts, points, and cash back show

that there is going to be a strong trend

towards emotional loyalty. Things like

recognition, relevance, and relationship

with the product. So to all founders,

techies, and users, of course,

gamification isn't dead. Yes, we're

absolutely bored of it at this point,

but it's about to change fundamentally.

So, say what you will about Gen Z

trends, but vibes are becoming the

factor when it comes to loyalty. As

always, we hope this was helpful. Till

next time. Bye.

衰落的原因:游戏化的崩溃

那么,为何它会走向衰败?那些最初激发兴奋感的心理诡计,最终成为了用户感到倦怠的根源。原本旨在让产品更有趣、使用起来更自然、并通过玩耍来帮助学习的游戏化(Gamification),最终沦为了商品化(Commodity)。

在游戏化应用盛行和野蛮生长的浪潮中,Gartner在2012年末曾做出预测:到2014年,80%的游戏化应用将以失败告终。其失败的原因在于糟糕的应用设计。这与我们当前目睹的AI泡沫,以及我们之前制作的关于“为何大多数AI初创公司不是好生意”的视频有着惊人的相似之处。因为,当缺乏扎实的产品作为支撑,仅仅追逐趋势是走不远的。

还记得我说过游戏化是不可避免的吗?智能手机、数据和注意力经济。这三个促成其必然性的条件,恰恰也是使其变得不可持续的因素。当iPhone于2007年问世时,App Store大约有500款应用。到了2015年,苹果和谷歌各自托管的应用数量已接近170万款,但人们实际使用的应用数量却仅维持在27款左右。而且,即便手机使用时长不断攀升,这一数字也趋于平稳。这意味着应用们在争夺一个人iPhone上的27个宝贵位置。在这样的环境下,游戏化成为了最廉价的用户留存(Retention)工具。积分、徽章、排行榜、连胜记录的实施成本几乎为零,相比之下,开发真正具有产品差异化的功能则需要巨大的投入。

问题在于,游戏化在短期内确实有效,但它依然有效。对企业而言,引入游戏化作为一种“打补丁”的修复方案是合乎逻辑的。以我多年的产品管理经验来看,有时你必须找到最经济的方式来解决问题。但这种方法是有代价的。对游戏化而言,它被轻易地“铆接”到应用上,成为一种留存手段(Retention Hack),而非构建一个拥有可防御价值、能抵御周期性炒作(Hype Cycles)的业务。

更糟糕的是,它还扭曲了“游戏”的本质,因为游戏并非关乎积分、金币或徽章。这些元素与游戏的距离,正如它们与健身或语言学习的距离一样遥远。连胜记录(Streaks)是现代游戏化中最有效的留存机制,但它们也带来了一个破坏性的次级效应:将真诚的兴趣转化为焦虑。即便是Duoingo也承认这一点。长期的连胜记录通过损失规避(Loss Aversion: 避免损失的心理偏好)来激励用户,而非积极的期待。这意味着,在学习一门语言时,连胜记录并不能让用户想要学得更多,反而让他们害怕停止,因此他们只会做最基本的事情来维持连胜。然而,维持连胜记录与学习本身毫无关联——连胜记录存在了,但你却几乎一无所获。

移动应用中学习的“游戏化”已经取代了产品的实际价值。Foursquare是游戏化疲劳最明显的商业案例。该应用在2013年拥有4500万用户,所有用户基础都建立在签到徽章和“市长”头衔的竞争之上。但到了2013年,签到的新奇感逐渐消退,游戏化无法维持用户的兴趣,因为该产品在游戏化之外的价值本身就不够充分。

那么,Duoingo为何没有走上同样的道路?

Original English

So why did it collapse? The

psychological tricks that sparked the

initial excitement became the very thing

that burned users out. Gamification that

was supposed to be a fun element that

would make the product less boring and

more organic to use and help you learn

things by playing became a commodity.

During the wave of fascination and wild

growth of gamified apps, Gartner made a

prediction in late 2012 that 80% of

gamified apps would fail by 2014. And

the reason for it was poor application

design. This is very similar to the AI

bubble that we're witnessing right now

and the video we made on why most AI

startups are bad businesses. Because you

can only get so far when pursuing trends

and not having a solid product. Remember

how I said that gamification was

inevitable? Smartphones, data, and

attention economy. These three

conditions that made it inevitable are

the very conditions that also made it

unsustainable. When the iPhone came out

in 2007, the App Store had about 500

apps. By 2015, Apple and Google each

hosted around 1.7 million, but the

number of apps people actually used was

around 27. And that number plateaued

even as time spent on phones kept

climbing. Which means that apps were

competing for 27 slots on a person's

iPhone. This environment made

gamification the cheapest available tool

for retaining users. Points, badges,

leaderboards, streaks cost next to

nothing to implement compared to

building genuine product

differentiation.

And the problem is it works in the short

term, but still it works. Introducing

gamification as a patch fix was rational

for businesses. Speaking from

experience, after building a career in

product management, sometimes you have

to find the cheapest possible way to fix

a problem. But this approach comes at a

cost. for gamification. It got bolted

onto apps as a retention hack rather

than building a business with a

defensible value that won't be wiped

away by hype cycles. And what's even

worse is that it disfigured the notion

of games because games are not about

points or coins or badges. These things

are as far from games as they are from

fitness or language learning. Streaks

are the most effective retention

mechanic in modern gamification. And

they also have a destructive second

order effect. They convert genuine

interest into anxiety. Even Dualingo

admits this. Long streaks motivate

through loss aversion rather than

positive anticipation. What this means

in playing English is that the streak

doesn't make users want to learn more.

It means that they're afraid to stop and

therefore they end up doing the absolute

minimum to simply preserve the streak.

But preserving the streak has nothing to

do with learning. The streak is there,

but you're learning essentially nothing.

The gamification of learning in mobile

apps has replaced the product's actual

value. Forsquare is the clearest

business case of gamification fatigue.

The app had 45 million users in 2013,

and all of that audience was built on

check-in badges and mayorship

competitions. But by 2013, the novelty

of checking in wore off, and the

gamification could not sustain the

interest because the value of the

product outside of gamification simply

wasn't there.

Duoingo的例外:精细化与数据驱动

Duoingo之所以成为一个值得称道的例外,是因为它投入了超过600项实验来精炼其游戏化机制,并最终“跑赢了跑步机”(outrun the treadmill)。无论我对Duoingo的学习机制持有何种看法,我都必须承认,他们成功构建了一个在发现和测试行为信号(Behavioral Signals)方面比任何竞争对手都更快速的系统。

行为经济学(Behavioral Economics)和心理学中,有一个术语叫做享乐适应(Hedonic Adaptation: 人类适应新刺激,快乐或不快乐感随时间减弱的现象)。Duoingo巧妙地管理着这一过程,它不断刷新视觉元素、规则、奖励以及“连胜冻结”(Streak Freezes)功能,以确保用户的大脑不会完全习惯于游戏机制。它们将应用程序打造成一个“鲜活的系统”(Living System)。

更重要的是,Duoingo构建了极其稳固且具有防御性的技术基础设施。他们追踪了数亿用户的微行为数据。在2010年代,大多数游戏化产品从未在如此深度上进行投入。即便放眼今天的市场,Duoingo的产品文化也属罕见。是的,它们确实是建立在积分和徽章之上,但它们同时也搭建了数据科学和实验的底层架构,用以深刻理解这些机制为何有效或失败。

从语言学习应用的角度来看——尽管可能有人会 disagree——我个人认为Duoingo的产品与语言学本身关联不大,但这仅仅是一个主观观点。然而,Duoingo无疑是一个值得关注的例外。大多数公司缺乏持续进行如此高强度实验所需的资源或文化,这也是Gartner预测的80%失败率在更广阔的市场中得到验证,但Duoingo却成功规避的原因。

AI经济下的未来:个性化与情感忠诚

那么,游戏化在AI经济时代将走向何方?尽管困难重重,游戏化市场预计将从200亿美元增长至2034年的1900亿美元。然而,八年后,所谓的“游戏化”将与其今天我们所见的积分和徽章系统截然不同。AI将对其进行极致的个性化处理,以至于每个应用程序内部都将内置一个“行为层”(Behavioral Layer)。

那么,具体会发生什么呢?

首先,个性化将达到用户级别。社交媒体、游戏、电子商务将能够读取用户的行为信号,精准地了解每个用户的动机。当用户即将流失时,AI知道如何即时调整体验,知道如何在通知中加入什么内容,如何调整课程难度,如何根据你今天的情绪状态改变你的训练计划。AI所做的,就是模仿Duoingo所开创的系统。AI会将许多产品转变为“鲜活的系统”,这些系统会不断变化和适应。

其次,请仔细聆听:忠诚度(Loyalty)的定义将被重塑。关于交易性忠诚(Transactional Loyalty)——即折扣、积分和现金返还——的最新研究和预测表明,一种强烈的趋势正指向情感忠诚(Emotional Loyalty)。这包括与产品相关的认可(Recognition)、相关性(Relevance)以及关系(Relationship)。

因此,对于所有创始人、技术人员以及用户来说,游戏化当然并未消亡。是的,我们此刻对此可能深感厌倦,但它即将发生根本性的变革。所以,无论你对Z世代的潮流有何看法,当谈及忠诚度时,“氛围感”(Vibes)正逐渐成为关键因素。

一如既往,希望本期内容对您有所帮助。下次再见。

Original English

Now, where is

gamification headed in the AI economy?

Despite all odds, the gamification

market is projected to grow from $20

billion to $190 billion by 2034. But the

term called gamification eight years

from now will bear no resemblance to the

points and badges system that we see

today. AI will personalize the hell out

of it to the point when there is going

to be a built-in behavioral layer inside

each app. So what exactly will happen?

Number one, personalization is going to

be on the user level. Social media,

gaming, e-commerce will be reading

behavioral signals and know exactly what

motivates each individual user. When

they're about to disengage, how to

change the experience on the spot, what

to put in your notifications, how to

adapt lesson difficulty, how to change

your training plan depending on how

you're feeling today. What AI will do is

mimic the system that Dolingo pioneered.

AI will turn lots of products into

living systems that will constantly

change and adapt. Number two, listen

closely. The definition of loyalty will

be reframed. Latest research and

predictions around transactional

loyalty. Transactional loyalty being

discounts, points, and cash back show

that there is going to be a strong trend

towards emotional loyalty. Things like

recognition, relevance, and relationship

with the product. So to all founders,

techies, and users, of course,

gamification isn't dead. Yes, we're

absolutely bored of it at this point,

but it's about to change fundamentally.

So, say what you will about Gen Z

trends, but vibes are becoming the

factor when it comes to loyalty. As

always, we hope this was helpful. Till

next time. Bye.

关键字: gamification behavioral-psychology ai-economy user-retention loyalty-mechanics