Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Validates ideas & hypotheses with A/B Testing using VWO.

Bung Momo
November 19, 2021

Validates ideas & hypotheses with A/B Testing using VWO.

Learning from how Netflix controls their user's decisions and validates hypotheses with A/B Testing using VWO.

Bung Momo

November 19, 2021
Tweet

More Decks by Bung Momo

Other Decks in Design

Transcript

  1. Did you know that Net ix thumbnails are important when

    choosing a movie or series to watch? ?
  2. Net ix creates a bunch of thumbnails for single movie

    or series... ... and shows you the best tailor- made thumbnail for you
  3. If you like old movies They'll probably show you this

    thumbnail, reffering to the movie "Ghostbusters". 👻
  4. If you like horror movies They'll probably show you something

    that ts with what you've watched before (thriller/horror) 👹
  5. And why this works? As you know, humans are visual

    creatures. Net ix found that the biggest in uence for a user to pick a movie were the thumbnails. 🎬 More personalization on the visuals equals highest chances to watch any movie or series (even if it's not the best for you)
  6. What if ... we implemented this feature on our platform?

    I think this way can make it easier for our users to choose a campaign to donate based on a personalized thumbnail. 🤔 To validate my hyphothesis let's run A/B Testing.
  7. What is A/B Testing A/B Testing also known as split

    testing is a method of comparing two versions of a webpage against each other to determine which one performs better. A/B Testing is an experiment for comparing two solutions. Its main purpose is to con rm a hypothesis. A/B testing is Quantitative Research, means the more users test it, the better results you get!
  8. Easy setup Shop all deals & offers Move / Resize

    Remove Inline Edit Copy Rearrange Select Parent Track Clicks DESIGN NAVIGATE Hide Edit Element Edit HTML SELECTION LIST Visual Editor Code Editor
  9. Quick Results , Control C 6.47% Blue Button V1 7.12%

    Green Button V2 5.57% Baseline Baseline 77% 10.00% 12% -13.93% 21% 75% 4% 124 / 1,926 109 / 1,539 104 / 1,877 Variations Conversion Rate Improvement Probability to beat Baseline Probability to be Best Conversions / Visitors Control C 6.47% Blue Button V1 7.12% Green Button V2 5.57% Baseline Baseline 77% 10.00% 12% -13.93% 21% 75% 4% 124 / 1,926 109 / 1,539 104 / 1,877 Variations Conversion Rate Improvement Probability to beat Baseline Probability to be Best Conversions / Visitors
  10. 01 02 03 Setting up pages for your test Specify

    the hypothesis, define the page URLs, and configure the visitor segments Creating variations using Visual Editor / Code Adding conversion goals Goal conversion when users click on a link, visit pages, submit forms and so on 04 Finalizing your test
  11. Control Var 1 What hypothesis are you testing? Thumbnails with

    non-disurbing image will address more attention & conversion Non-disturbing image Disturbing image Urgent Butuh Operasi Karena Tak Bisa BAB https://kitabisa.com/campaign/bantuoperasianus C V1
  12. 01 02 03 Test ideas and hypotheses anytime Ease of

    analysis Everything is testable 04 Quick results