Form Analytics: Conversion rate as unweighted average?

  • harmerschau
    Asked on December 6, 2017 at 12:02 PM

    Hi.

    I wondered about the conversion rate for this form

    https://www.jotform.com/analytics/73226626006148

    with 3 responses over 67 views and a "conversion rate" of 16% (vs. 3/67=4% I would expect).

    This answer: https://www.jotform.com/answers/849722-How-is-conversion-rate-calculated

    says that the conversion rate is calculated as the (unweighted) average of the daily conversion rates and indeed when I add the daily conversion rates and divide by the number of days, I get 16%.

    But why calculate the overall conversion rate this way?

    Imagine that I have a form up for two days. I test it on the first day and fill it out, and then release it and get responses on the second day.

    Day     Views    Responses    Daily Conversion Rate

    1            5             5                   100%

    2          100            4                      4%


    As I understand it, the algorithm will calculate an overall "conversion rate" of 52% (=104%/2).

    That's awesome, but really doesn't reflect the form's performance, which is better given by the weighted average of 9/105 = 9%.

    This would be equally uninformative, indeed deceptive, in the other direction:

    Day     Views    Responses    Daily Conversion Rate

    1            5             0                      0%

    2            5             0                       0%

    3            5             0                       0%

    4          100           40                     40%

    where the jotform "conversion rate" of 10% vastly understates how well the form performs when it gets traffic. 

    The weighted average conversion rate gives me a basis for estimating, for instance, "If I get 1000 views tomorrow, how many responses should I expect?" What useful question does the unweighted average conversion rate answer for me?

    As this clever person says, "Taking average of percentages (what you call "unweighted" average) will give you useless results if your samples differ in size."

    Why waste dashboard space (and mislead naive users) with this useless calculation?




  • BJoanna
    Replied on December 6, 2017 at 1:05 PM

    I am not sure why the "conversion rate" is calculated that way. I will have to check this with our developers. I will also check with them if we can change the way the "conversion rate" is calculated or add another "conversion rate" field that will calculate the average (views/responses).

    If there is any update we will inform you via this thread. 

  • harmerschau
    Replied on December 8, 2017 at 1:18 PM

    Thanks for your reply.

    I think it would be a very bad idea to stick with the current conversion ratio, but if you do, a FAQ on how the unweighted average is meaningful and what it can be used for is a must. 

  • BJoanna
    Replied on December 8, 2017 at 2:33 PM

    I raised your request to our developers. However, I do not know if we will change the way the "conversion rate" is calculated or not. We will inform you via this thread if there is any update regarding your request.