How to A/B Test Emails | Ecommerce Examples | Best Practices

Поділитися
Вставка
  • Опубліковано 5 лис 2024

КОМЕНТАРІ • 19

  • @SamTechWorld9
    @SamTechWorld9 2 роки тому +3

    I wish you were my email marketing teacher, Id pay attention to every single word. Great content!

    • @KaseyLuck
      @KaseyLuck  2 роки тому +1

      Thank you! Feel free to use my UA-cam channel as the email marketing school :)

  • @shivamk1447
    @shivamk1447 Рік тому +6

    Loved the video. So informative.
    One question: When you decide Test pool size, how do they get allocated for Test groups (Test group A and Test group B)?
    Example you have given, out of 10k, 3k will be the test group. So are both Test groups A and B divided 15% each and winning criteria will automatically will be sent to the 70% non-test group?

    • @KaseyLuck
      @KaseyLuck  Рік тому +2

      yes, that's exactly correct. all that setup should happen automatically

  • @lucalotzbeyer6823
    @lucalotzbeyer6823 2 роки тому +3

    Hey Kasey, you're speaking of bigger Subscriber sizes of at least 1k+ People . But what if the subscriber size is really small, let's say 50 people only. Do you think that it's even worth to A/B-Test at that stage? Or do you think it isn't worth the effort because there is no statistical evidence due to the small size? Love your videos and would also love a quick response!😀

    • @KaseyLuck
      @KaseyLuck  2 роки тому +6

      hey Luca! With an audience size of below 1,000 subscribers, A/B testing is not worth the effort because there is no statistical evidence due to the small size, as you've rightfully mentioned. I wouldn't worry too much about A/B testing until you're at about 3,000 or even 5,000 subscribers. Your attention would be better spent at growing your audience and creating great content for your subscribers. With a smaller audience, you could do "informal" A/B testing where you're not technically splitting the audience, but you're sending different types of content and see which topics resonate best over time.

    • @lucalotzbeyer6823
      @lucalotzbeyer6823 2 роки тому

      @@KaseyLuck Ok, got it. Thank you so much for your detailed answer! I'll definitely try the content testing :)

  • @saraa404
    @saraa404 Рік тому

    nice video. thanks. I have a question: the 30% of the email receivers might take time to open their email so would the 70% of the people who are supposed to get a winning variant of the test email, would have to wait till the result compiles?

    • @KaseyLuck
      @KaseyLuck  Рік тому

      yep! and you get to choose the number of hours you want to wait before sending the winning variation to the remaining 70% of your audience

    • @saraa404
      @saraa404 Рік тому +1

      @@KaseyLuck thanks

  • @tugcemayaturk2494
    @tugcemayaturk2494 2 роки тому

    Hey Kasey ! LOVE your videos. I would love to work with you in the future. Love your energy!

    • @KaseyLuck
      @KaseyLuck  2 роки тому

      Thank you Tugce!! 🙌

  • @JAHANGIRALAM-ik4mj
    @JAHANGIRALAM-ik4mj Рік тому

    beautiful

  • @zakariaabali4750
    @zakariaabali4750 Рік тому

    Who told you, you cant track open rates, if you use a crm theres a window to track it

    • @KaseyLuck
      @KaseyLuck  Рік тому +1

      you can track it, but the tracking is not reliable since all emails sent to users on iOS 15+ are counted as opened.

    • @zakariaabali4750
      @zakariaabali4750 Рік тому

      @@KaseyLuck nah dear, crm have their banchmarking tools where u can track open rate, click rates.. everything just double check

    • @zakariaabali4750
      @zakariaabali4750 Рік тому

      Just check Klaviyo! I still can see that emails arent opened while other are open and they are accurate

    • @KaseyLuck
      @KaseyLuck  Рік тому +1

      @@zakariaabali4750 are you familiar with the iOS15 update? I talk more about it in this video: ua-cam.com/video/zmY0LK3Wu7Q/v-deo.html

    • @zakariaabali4750
      @zakariaabali4750 Рік тому

      @@KaseyLuck alright, now you think we have to focus on click throught rate instead