Project SmartKitchen/ Experiment 3: Facebook Ad

From MozillaWiki
Jump to: navigation, search

Status: DONE

Leap of Faith Assumption (LOFA) being tested:

  • (S) We assume that people want recipe ideas that they can make with their existing inventory


If run a Facebook Ad describing our product, we expect a 2% response rate

Experiment Overview:

  • "We ran a Facebook Ad ""Wondering What to Cook"" on the rightside column (ad section) to US, 18+, men and women, interested in cooking:recipes (TAM of 13M)
  • The ad was optimized for website clicks with a $50 budget spent between 2:30pm and 9:pm on July 27, 2016. Clicking led to a surveygizmo site explaining the features (see details and notes) and asked users to take a survey in lieu of collecting email addresses (not cleared by legal)."

Experiment Timeline:

Duration: 7.5 hours, July 27, 2016 from 2:30pm to 9:pm


  • FB clickthrough - tests really the ad and 90 character message, but still useful
  • Survey results - percent of people who complete survey, percent of people selecting each problem statement

Experiment Details and Notes:

  • "In this experiment we are trying to validate a user problem. That is that people want suggestions to use the ingredient that they have on hand. We are testing these things:

    • Save Time - reduce the amount of time you spend planning and stressing out about what to cook, making planning fun instead of a chore.
    • Discover New Recipe Ideas - get imaginative recipe suggestions that use the ingredients that you already have on hand.
    • Satisfy Picky Eaters and Special Dietary Requirements - receive tailored suggestions to satisfy all your household dietary requirements.
    • Reduce Food Waste - reduce the amount of food you throw away because of spoilage or expiration.
    • Remember What to Shop For - receive reminders of what to shop for or commonly used items you've run out of.

Facebook Ad Screenshots:
Smartkitchen facebook experiment.png


Smartkitchen facebook results.png

  • We received 0 survey results and poor response to the ad.
    • Next Step: See what the problem is with the ad, we are not really optimizes for the ad, but want to get results in the survey.
  • .22% response rate on Facebook Ad
  • "Stir Fry" performed better.
  • 55 website clicks
  • 0 survey results
  • Poor results
  • Women were 80% of respondents.

Discussion on Results:

  • We got poor facebook response because it was placed in the right hand side advertising area (who clicks on these?), or had poor wording, or had to expose the URL "" and that may reduce response rate
  • NEXT STEP: see if we can increase FB response rate by creating a sponsored ad in the news feed, and changing the copy for the ad.
  • Next experiment: Will develop a facebook page for smartkitchen, and do a split response testing different messages but holding picture constant right now on "Stir Fry"