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

The Science of Ranking Features

Gilles Bertaux
February 10, 2014

The Science of Ranking Features

As a product manager or designer, when it comes to building a website, an app, or any product for that matter, it could be really difficult to establish the features you want to build first and define a roadmap.

Recently, I came up with a small process that I use in my own projects to attribute scores on defined features in order to rank them and maximise productivity.

Please, let me know if you have any idea to improve this method.

Gilles Bertaux

February 10, 2014
Tweet

More Decks by Gilles Bertaux

Other Decks in Design

Transcript

  1. The Science of Ranking
    Features
    Product managers, here’s a homemade process to build
    features that work for your audience.

    View full-size slide

  2. Defining and ranking features
    lead to anger, anger leads to
    hate, hate leads to suffering.

    View full-size slide

  3. « That killer-feature-we-are-
    working-on-for-8-months HAS
    to be built. Because it’s cool. »

    View full-size slide

  4. This is bad and you
    should feel bad.

    View full-size slide

  5. The path to success is
    a two factor
    formula…

    View full-size slide

  6. …And user needs.

    View full-size slide

  7. Speaking of user
    needs, do you know
    the Kano Model ?

    View full-size slide

  8. 4 Principles
    • Un : User need score from 0 to 10
    • t : time factor (1, 2, etc. but no 0.5)
    • Final score will be defined by Un/t
    • The closer to 0, the less important

    View full-size slide

  9. Defining Un
    Feedback is essential. I counts for 5
    points out of 5.

    View full-size slide

  10. Defining Un - Combo
    coeficients
    • “Must Be” = 2.
    • “Delighter” = 1.
    • “Perfomance” = 1.5.

    View full-size slide

  11. Let’s put it to
    practice now.

    View full-size slide

  12. Consider a Facebook
    Login feature.

    View full-size slide

  13. My users rated it
    2.5/5.

    View full-size slide

  14. We considered it as a
    performance
    attribute.

    View full-size slide

  15. Therefore, Un =
    2.5*1.5 = 3.75.

    View full-size slide

  16. Now, let’s say we can
    built it properly in 2
    hours.

    View full-size slide

  17. The feature get the
    final score of 3.75/2
    = 1.875

    View full-size slide

  18. There you have a
    simple, light process
    to organize your
    feature.

    View full-size slide

  19. The Science of Ranking
    Features
    This article was originally built on Medium : http://
    medium.com/p/613f89d2c6b6

    View full-size slide

  20. Gilles Bertaux
    http://gillesbertaux.com
    @Gilles_Bertaux
    About the author

    View full-size slide

  21. Font : Myriad Pro
    Images: Breaking Bad, Star Wars, A Serious Man, Basic
    Instinct, Hunger Games, The Time Machine, Seven
    Samurais.
    Credits

    View full-size slide