回复: 8
跳转到指定楼层
上一主题 下一主题
收起左侧

分享遇到的两个很好的ab testing的题目和suggested answers!

   
全局:

2022(10-12月) 分析|数据科学类 硕士 全职@google - 内推 - HR筛选 Onsite  | 😃 Positive 😐 Average | Other | 应届毕业生

注册一亩三分地论坛,查看更多干货!

您需要 登录 才可以下载或查看附件。没有帐号?注册账号

x
感觉很经典的两道题目!
1、Testing Price Increase
Let’s say that you work at a B2B SAAS company that’s interested in testing the pricing of different levels of subscriptions..--
Your project manager comes to you and asks you to run a two-week-long A/B test to test an increase in pricing.. Χ
How would you approach designing this test? How would you determine whether the increase in pricing is a good business decision?
Let’s tackle one overarching question first - should we run an A/B test when testing pricing?.1point3acres

answers:A/B testing pricing generally has more downsides than upsides. One major downside is that if two users go to a pricing page and one of them sees a product for \$50$50/month and the other sees one for \$25$25/month, you’re risking an incentive for your users to opt-out of your A/B test into another bucket, creating real statistical anomalies.
But an even larger issue on A/B testing pricing is on understanding success? If you’re testing a discount rate on a subscription product - you want to know two things:
1.        Does the customer convert at a higher rate for the discount?. check 1point3acres for more.
您好!
本帖隐藏的内容需要积分高于 188 才可浏览
您当前积分为 0。
使用VIP即刻解锁阅读权限或查看其他获取积分的方式
游客,您好!
本帖隐藏的内容需要积分高于 188 才可浏览
您当前积分为 0。
VIP即刻解锁阅读权限 或 查看其他获取积分的方式
Unlock interview details and practice with AI
Curated Interview Questions from Top Companies
would be a greedy version of running the test given that if the red button won and then the bottom of the page variant won, we wouldn’t have known if the interaction effect of blue + bottom of the page may have changed the outcome.
Let’s make sure to add into the solution to note the general design of the AB test as well to not get caught up in the specifics of the multivariate testing. We need to calculate the sample size we would need by multiplying the number of users per day that would be bucketed in each variant by the number of days that it would take to reach significance. We also need to note that we would randomly bucket each user into each variant in order to not bias the results.


求rice看面积哇,谢谢各位大佬!

补充内容 (2022-11-06 13:04 +8:00):.
sorry,面经

评分

参与人数 8大米 +29 收起 理由
匿名用户-0TH1A + 22
margowang + 1 赞一个!
ChampRacer + 1 赞一个
小亩_b0e57d5 + 1 很有用的信息!
做一条咸鱼 + 1 楼主/层主请继续!

查看全部评分


上一篇:Tiktok ds ng面试内容求指教
下一篇:家具网 ds/ml intern öâ codility
推荐
lllxxy 2025-2-27 09:27:33 | 只看该作者
全局:
谢谢分享,第二题不是应该用 ANOVA 来分析吗?
回复

使用道具 举报

全局:
谢谢分享!
回复

使用道具 举报

🔗
lagender 2024-4-20 10:29:34 | 只看该作者
全局:
谢谢楼主分享!!
回复

使用道具 举报

🔗
supertankaa 2024-12-9 03:43:23 | 只看该作者
全局:
谢谢楼主分享
回复

使用道具 举报

全局:
谢谢分享
回复

使用道具 举报

地里匿名用户
🔗
匿名用户-IJO8W  2025-1-18 14:45:49
感谢分享
回复

使用道具 举报

🔗
joegu 2025-3-15 02:26:02 | 只看该作者
全局:
xie xie fen xiang
回复

使用道具 举报

全局:
谢谢分享
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 注册账号
隐私提醒:
  • ☑ 禁止发布广告,拉群,贴个人联系方式:找人请去🔗同学同事飞友,拉群请去🔗拉群结伴,广告请去🔗跳蚤市场,和 🔗租房广告|找室友
  • ☑ 论坛内容在发帖 30 分钟内可以编辑,过后则不能删帖。为防止被骚扰甚至人肉,不要公开留微信等联系方式,如有需求请以论坛私信方式发送。
  • ☑ 干货版块可免费使用 🔗超级匿名:面经(美国面经、中国面经、数科面经、PM面经),抖包袱(美国、中国)和录取汇报、定位选校版
  • ☑ 查阅全站 🔗各种匿名方法

本版积分规则

>
快速回复 返回顶部 返回列表