Forecasting best practices

jeaninejiang
jeaninejiang Contributor
edited March 13 in Datasets

Hi could you please share some best practices of forecasting?

Best Answer

  • jaredpk
    jaredpk Domo Employee
    Answer ✓

    At a high level, here are a couple of tips that I find helpful.

     

    First, look at your actual transactional data to see what dimensions are important to you.  Once you determine the actual dimensions that are of value, create a forecast sheet that includes those dimensions.  Just remember, the more detailed you are with dimensions the more lines of data will be required in your forecast.  However an hour or two of pre-work when setting the forecast can allow for much more detailed tracking throughout the year.

     

    Second, make sure the metric and dimension names in the forecast align with those in your actual data.  Without that, it is much harder to set up cards to show comparisons.

     

    Third, we often place the forcecast data in a Domo online form, so it can be accessed and updated as needed.

     

    Lastly, coming up with the forecast numbers is a bit of art and science.  Obviously a good place to start is with prior history.  But internal initiatives, market conditions, and other changes will affect future performance.

     

    If you have more specific questions, feel free to post.

    I work for Domo.
    **Say "Thanks" by clicking the thumbs up in the post that helped you.
    **Please mark the post that solves your problem as "Accepted Solution"

Answers

  • nalbright
    nalbright Contributor

    Did you have any particular aspects of forecasting you were most curious about?

    "When I have money, I buy books. When I have no money, I buy food." - Erasmus of Rotterdam
  • What are you using as your CRM?

  • jaredpk
    jaredpk Domo Employee
    Answer ✓

    At a high level, here are a couple of tips that I find helpful.

     

    First, look at your actual transactional data to see what dimensions are important to you.  Once you determine the actual dimensions that are of value, create a forecast sheet that includes those dimensions.  Just remember, the more detailed you are with dimensions the more lines of data will be required in your forecast.  However an hour or two of pre-work when setting the forecast can allow for much more detailed tracking throughout the year.

     

    Second, make sure the metric and dimension names in the forecast align with those in your actual data.  Without that, it is much harder to set up cards to show comparisons.

     

    Third, we often place the forcecast data in a Domo online form, so it can be accessed and updated as needed.

     

    Lastly, coming up with the forecast numbers is a bit of art and science.  Obviously a good place to start is with prior history.  But internal initiatives, market conditions, and other changes will affect future performance.

     

    If you have more specific questions, feel free to post.

    I work for Domo.
    **Say "Thanks" by clicking the thumbs up in the post that helped you.
    **Please mark the post that solves your problem as "Accepted Solution"
  • @jeaninejiang did any of the above replies help you? If so please mark the correct reply by clicking on "Accept as Solution".

     

    Thanks!

    Dani

  • the replies are helpful, but is there a location to see examples or use cases to use as references?