Skip to content

Feature Experimentation

Categories

JUMP TO ANOTHER FORUM

7 results found

  1. Clients wants to be able to have an approval workflow before a flag or test is started. They do this because they want to ensure that someone needs to get approval before changing the configuration of a flag to ensure a second pair of eyes before changes are made (similar to approving code changes.

    4 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  2. As a developer I want to be able to use an Optimizely provider for the CNCF OpenFeature project so that I don't have to learn a vendor-specific SDK to integrate my codebase with Optimizely resulting in faster adoption of Optimizely in my applications

    3 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  3. requester chase@weedmaps.com

    Type of issue: Product feedback
    Chase Dutton sent a feedback.
    Description: When you guys launched the new UI, it confused the heck outta every developer at Weedmaps and we had many product incidents where a flag was misidentified as being off when it should have been on, or vice versa.
    This confusion wasn't due to any bugs in Optimizely or Weedmaps, just the new UI was visually confusing all who encountered it.
    Anyways we're doing much better now, but yea just wanted to share that feedback. If it confused 100% of our 60+ devs I'm sure it confused…

    2 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  4. requester ivan.njunjic@aura.com

    Type of issue: Feature Experimentation
    Ivan Njunjic sent a feedback.
    Description: I'd like to measure the effectiveness of our experimentation program through a quarter-long holdout experiment. Is there an easy, self-serve method to set up holdout groups (say, 5% of traffic that would not receive any experiment treatments during the quarter, or do we need to manage this in code via a separate flag? Thanks!
    User email: ivan.njunjic@aura.com
    User id: 3y-jacmkR2mDxmoPks63Vg
    Account id: 7e76915c-e202-a8ec-da88-eda5848b9576
    http://notify.aptrinsic.com/wf/open?upn=Q6elNXBfOd1okuIEgGFEjVYK-2FQeJKmhJxBg4Gdoz-2BpoHYFLIZK2l4Z3cLM28FjN48rQtYKJlQIVk6MrNwrilTz3oceTf5cdas3ShflIPI1NBLe8f3Ld5DuAwwhKHl3EazDUY6TP0JiEFq9GTDqeeJAT26dbbGSrZnzvQKBlYSviT35S8fQKPcHicRTQoGa4PU2MCDHNoxB-2BDxYsgBTJVnmqJ0NUeW43gFOgy0-2FNwQpr6bu1L3nCReD03QxBYZGIP

    2 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  5. In Flags, it is not possible to change the order of variants and have a different "Baseline" variation by default in the Results Page.
    Regardless the possible workaround (change the baseline variation from the results page, the proper behaviour would be to being able to switch the order of the variants (and so decide what the default "Baseline" variant is directly in the Flags Page.

    2 votes

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  6. While the decision object allows you to see which rule delivered a feature, it does not specify whether this rule is an A/B Test, Targeted Delivery, etc.
    Customer naming conventions could theoretically provide this kind of context, but we've received a request from Nike to make this available in the SDK itself.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  7. While the decision object returned from the SDK currently provides the flag status, variation key, and rule key, there is currently no built-in way to determine whether the rule that returned the decision was an A/B test or targeted delivery. A workaround would be to use specific naming conventions for different rule types, but it would be useful to have this information baked into the decision object by default.

    1 vote

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)

    We’ll send you updates on this idea

    How important is this to you?

    We're glad you're here

    Please sign in to leave feedback

    Signed in as (Sign out)
  • Don't see your idea?