Growth Hacking精要-跟Airbnb、Pinterest和Uber学习增长秘笈

原作者: Charu Angid [原文链接]

Earlier tonight, I attended a growth tech talk held at Airbnb. I know quite a few people who were on the waitlist for this event and some who were not able to attend in person, so here’s my summary of key learnings from the talks. I’m going to focus on things that I found unique to each company. You can also find a full recording of the talks on Airbnb’s website here.

今晚华灯初上之时,我在Airbnb参加了一个关于增长技术的分享活动。我知道很多人在这次活动的候补名单上,而有些人自己无法到场,所以我就把对这次talk的总结放在这里以飨大家。总结中我将聚焦在每家公司不一样的地方,你可以在Airbnb的网站上找到他们所有活动的录音

If I had to sum up the key growth strategy themes for these 3 companies in a few words (based on the non-statistically significant # of data points I collected on each of these companies from the talks) it would be the following ->
Airbnb: Carefully crafted quality user experience
Pinterest: Discovery of pinnable and pinned content
Uber: Rapid scalable deployment across the world

基于活动中个人对这几家公司重要观点的衡量,它们各自增长策略的精要大致如下:
Airbnb: 精心雕琢高质量用户体验
Pinterest: 发现值得钉上和已经钉上的图片内容
Uber: 全球市场的快速规模化复制

While I’m 100% certain all 3 of these companies care about all 3 of these things and more (hypothesize, test, iterate) when it comes to growth, the reason I picked these as top level themes were based on the way the speakers at these companies communicated the problems that they thought about the most, and how they approached solving them.

虽然我100%确定这三家都关心以上三个指标和其他增长黑客技术(假设、测试、迭代),我选择以上几点作为最重要主题的理由是每家公司发言人在分享这些核心问题时的表达方式,以及他们解决这些问题的方法。

For Airbnb, growth is focused on 3 main goals: building a quality experience, having no negative effect on the ecosystem, and driving 40% lift in new users. Chris, Vivek and Nick from the Airbnb growth team shared concrete examples of tests they ran and how they made decisions to ramp/not ramp these changes. For instance, Airbnb tested reducing the frequency with which they showed a sign up prompt asking users to create an Airbnb account or log in - while reducing the frequency of showing this modal reduced signups by 4%, the quality of user experience improved, which led the team to commit to the change. In another example, the team removed testimonials from the sign up modal, which led to a 9% increase in sign ups, without any observable negative user cost. Finally, the team discussed a test they ran where they increased the size of the sign up / log in modal and made copy changes that indicated continuity. These changes, which might be considered minor design tweaks, led to a significant increase in sign ups. The key takeaway from the Airbnb talk was that user experience and design play a big role in determining what types of tradeoffs to make and also drive growth features.

对于Airbnb, 增长主要集中在三个主要目标打造高质量体验,生态系统中无任何负面影响,并驱动40%的新客增长率。 来自Airbnb增长团队的ChrisVivekNick分享了他们测试的具体实例以及他们如何决策是否大肆推广这些变化。比如,Airbnb测试过减少弹出用户注册或登录账户的注册窗口提示的频率,结果显示,尽管新用户注册率下降了4%,用户体验的质量却上升了,这坚定了团队减少以上提示的决心。在另外一个例子里,团队将自我证言从注册流程中去掉,这提升了9%的注册率却没观察到任何对客户的负面影响。最后,团队分享了另一个例子,他们扩大了注册/登录窗口的尺寸并复制变化来提示连贯性。这些变化虽然可被认为设计上的微调,却大大地提高了注册用户数。Airbnb的分享中,我体会最深的就是:用户体验和设计在团队决定做出哪些类型的折衷且能驱动增长特征时居功至伟。

For Pinterest, SEO is a key driver of traffic to content. Vadim from Pinterest discussed a key tradeoff that SEO growth strategists often have to make - should you be optimizing for search engine crawlers to find your content or should you optimize for the user experience of real people who are actually consuming the content? For Pinterest, the focus is the latter. An example of a test Pinterest ran recently, was changing how pinned boards are ordered from chronological order to ordering by popularity. The hypothesis was that the new order would perform better as it had more popular pins. However, results showed a 20% drop in traffic. After investigations, the team found that the reason behind this was that popular pins have been repinned so many times, that each pin takes up more space in the user’s screen - resulting in fewer pins being surfaced to users which led to the traffic drop. Vadim also discussed how Pinterest optimizes for traffic from image search by using the original image searched for as an anchor and showing related pins for discovery of Pinterest content. The key takeaway from the Pinterest talk was that enabling discovery of pins and pinnable content in the best way is a core growth focus that leads to engagement and value for Pinterest users.

对于Pinterest,SEO是内容流量增长的关键抓手。 来自P家的Vadim分享了SEO增长专家常常要做的一个关键的折衷选择--是优化搜索引擎爬虫来找到你的内容还是该优化那帮真正消化内容的用户的体验?对于Pinterest,重点在于后者。Pinterest最近的一个例子就是,将被钉住的图片面板的显示顺序从按时间排序变成按流行度排序。其背后的假设就是新的顺序会更给力,因为按照流行度的图片面板上有更多的受欢迎的图片。然而,结果显示有20%的流量下降。经调查发现,那些受欢迎的图片因为已经被反复钉过很多次,每个钉都会占用屏幕上更多的空间,结果就导致更少的钉被显示到用户屏幕,从而产生了流量下滑。
Vadim也讨论了P家是如何优化图片搜索带来的流量峰的,先是通过锁定对原始图片的搜索,然后再显示Pinterest中与之相关的图片的钉。
Pinterest分享的关键点:使钉住的图片和值得钉住的内容的发现以最好方式展现,这是增长的核心所在,也是带来用户黏性和为用户创造价值的关键所在。

For Uber, growing fast in markets across the world, especially in large emerging economies like China and India is key. John and Anant from Uber shared strategies the engineering team deploys to facilitate rapid scalable growth. Uber has built an ‘onboarding state machine’ - a backend system for onboarding the driver experience with APIs, which is used in combination with country specific logic and UI modules, to deploy quickly in new markets that Uber enters. The system has gotten to a point where it can be used in a plug and play format with minimal refactoring. Uber has also developed sophisticated data science models which enable them to learn from the first few interactions users in new markets take on the site/app and predict if they will convert, which helps inform what pieces of the onboarding experience are working. The key takeaway from the Uber talk was that engineering for rapid deployment can be a massive advantage in entering new markets and quickly getting to product-market fit.

对于Uber,在全球市场快速增长,特别是在大型新兴经济实体(如中国、印度)的布局是关键。 来自U家的JohnAnant分享了他们开发团队用来推动快速规模化的策略。Uber打造了一个“适职状态机”--一个通过API来管理司机体验的后端系统,它通过和不同国家特有的逻辑和用户交互模块协同工作,从而快速地在Uber新进入市场上得以应用。该系统已经达到了仅需最小程度的重构便能即插即用的地步。U家同事也开发了复杂的数据模型来让他们从新市场的第一批种子用户身上学习并预测他们是否会变成Uber用户,这有助于Uber了解适职状态机中哪些模块在起作用。U家分享给我的最大启发就是:技术驱动和快速开发应用是进入新市场后打开局面和迅速找到市场契合点时的巨大优势。

Overall, I found the content of the talks to be highly relevant to the work we do every day at fast growing tech companies. +1 to the Airbnb team for throwing another great event.

好了,总而言之,我发现这些分享的内容和我们这些快速增长的科技公司里的日常工作息息相关。谢谢Airbnb团队主办了一次这么赞的活动。

最后编辑于
©著作权归作者所有,转载或内容合作请联系作者
  • 序言:七十年代末,一起剥皮案震惊了整个滨河市,随后出现的几起案子,更是在滨河造成了极大的恐慌,老刑警刘岩,带你破解...
    沈念sama阅读 213,254评论 6 492
  • 序言:滨河连续发生了三起死亡事件,死亡现场离奇诡异,居然都是意外死亡,警方通过查阅死者的电脑和手机,发现死者居然都...
    沈念sama阅读 90,875评论 3 387
  • 文/潘晓璐 我一进店门,熙熙楼的掌柜王于贵愁眉苦脸地迎上来,“玉大人,你说我怎么就摊上这事。” “怎么了?”我有些...
    开封第一讲书人阅读 158,682评论 0 348
  • 文/不坏的土叔 我叫张陵,是天一观的道长。 经常有香客问我,道长,这世上最难降的妖魔是什么? 我笑而不...
    开封第一讲书人阅读 56,896评论 1 285
  • 正文 为了忘掉前任,我火速办了婚礼,结果婚礼上,老公的妹妹穿的比我还像新娘。我一直安慰自己,他们只是感情好,可当我...
    茶点故事阅读 66,015评论 6 385
  • 文/花漫 我一把揭开白布。 她就那样静静地躺着,像睡着了一般。 火红的嫁衣衬着肌肤如雪。 梳的纹丝不乱的头发上,一...
    开封第一讲书人阅读 50,152评论 1 291
  • 那天,我揣着相机与录音,去河边找鬼。 笑死,一个胖子当着我的面吹牛,可吹牛的内容都是我干的。 我是一名探鬼主播,决...
    沈念sama阅读 39,208评论 3 412
  • 文/苍兰香墨 我猛地睁开眼,长吁一口气:“原来是场噩梦啊……” “哼!你这毒妇竟也来了?” 一声冷哼从身侧响起,我...
    开封第一讲书人阅读 37,962评论 0 268
  • 序言:老挝万荣一对情侣失踪,失踪者是张志新(化名)和其女友刘颖,没想到半个月后,有当地人在树林里发现了一具尸体,经...
    沈念sama阅读 44,388评论 1 304
  • 正文 独居荒郊野岭守林人离奇死亡,尸身上长有42处带血的脓包…… 初始之章·张勋 以下内容为张勋视角 年9月15日...
    茶点故事阅读 36,700评论 2 327
  • 正文 我和宋清朗相恋三年,在试婚纱的时候发现自己被绿了。 大学时的朋友给我发了我未婚夫和他白月光在一起吃饭的照片。...
    茶点故事阅读 38,867评论 1 341
  • 序言:一个原本活蹦乱跳的男人离奇死亡,死状恐怖,灵堂内的尸体忽然破棺而出,到底是诈尸还是另有隐情,我是刑警宁泽,带...
    沈念sama阅读 34,551评论 4 335
  • 正文 年R本政府宣布,位于F岛的核电站,受9级特大地震影响,放射性物质发生泄漏。R本人自食恶果不足惜,却给世界环境...
    茶点故事阅读 40,186评论 3 317
  • 文/蒙蒙 一、第九天 我趴在偏房一处隐蔽的房顶上张望。 院中可真热闹,春花似锦、人声如沸。这庄子的主人今日做“春日...
    开封第一讲书人阅读 30,901评论 0 21
  • 文/苍兰香墨 我抬头看了看天上的太阳。三九已至,却和暖如春,着一层夹袄步出监牢的瞬间,已是汗流浃背。 一阵脚步声响...
    开封第一讲书人阅读 32,142评论 1 267
  • 我被黑心中介骗来泰国打工, 没想到刚下飞机就差点儿被人妖公主榨干…… 1. 我叫王不留,地道东北人。 一个月前我还...
    沈念sama阅读 46,689评论 2 362
  • 正文 我出身青楼,却偏偏与公主长得像,于是被迫代替她去往敌国和亲。 传闻我的和亲对象是个残疾皇子,可洞房花烛夜当晚...
    茶点故事阅读 43,757评论 2 351

推荐阅读更多精彩内容

  • 今晚睡不着,忽地想起一名朋友来,是我的高中舍友,名叫邓泽,大学的几年中,我们几个投的来的高中舍友一直在群里玩这耍那...
    AJI米阅读 771评论 0 0
  • 人类一直都渴望能够飞翔,没有翅膀,所以发明了飞机。视觉是有限的,但是心灵是无限的。 人们喜爱鸟类,那些美妙的瞬间就...
    平凡的我和我阅读 267评论 0 1
  • Contiki内核是基于事件驱动和Protothreads机制,事件既可以是外部事件(比如按键,数据到达),也可以...
    zplodge阅读 1,138评论 0 0