Surveys and Outreach

Get Involved. Join the Conversation.

Topic

    Pavol Procka
    Agent Survey ReportingAnswered
    Topic posted November 21, 2017 by Pavol ProckaSilver Crown: 22,500+ Points 
    197 Views, 4 Comments
    Title:
    Agent Survey Reporting
    Content:

    Hi,

     

    I am trying to report on all Surveys that were sent for a specific agent no matter whether they were responded to.

    The reason for this is that we have 2 agents who are not receiving any QA survey results. When I looked through a few of their email incidents  manually I could see that the QA survey function seems to be firing. I would like to confirm whether it is firing on all of their emails.

    The 2 agents are using the same profile, group, workspace as the other 25 for which it works fine.

    Any pointers please?

    Many Thanks

    Pavol

    Best Comment

    Jess Campbell

    Hi,

    Our integrator created a custom field for us called survey sent. When the survey is triggered by business rules, the incident custom field survey sent is set to yes. This is a super easy way to identify for which incidents a survey was sent. (Easier than combing through transactions for each incident.)

    If you want to look for the info by transaction, the table association will be incident, the transaction type will be 24 (Sent), you can transform attribute1 into readable data by using the following formula: if(transactions.id1>0, msg_lookup(31927)), which should populate the column with Survey Sent. Description will show the email address to which the survey was sent. You will want to severely limit the filters to pull only this one transaction for any incident which you want to verify.  Transaction Type= 24, Transaction Table = Incidents, Transaction Attribute 1 >0, Incident ID in list (your list of email incidents for these two agents). I would choose a sample of them, as too many incidents combined with the transaction table being ridiculously large will make your report not run.

    You may also want to consider whether there are other conditions on the survey trigger (like opt-ins, or recency, or frequency) which are stopping some from being sent. Additionally, we have found that some groups of our customers are more likely than others to respond, so if the agents in question are more likely to handle a different product line or type of incident than other agents in the same profile, you may see a reduced response rate based on that as well.

    Just some things to consider. Good luck!

     

    Comment

     

    • Carl Elliott

      I think you could modify the Incident Detail by Survey report #7042 to get what you need.   Take out the responses and get it down to the survey and the incident and you could count the # of Incidents or transactions by the agent assigned to the incident.  Looking at the 'sent' transaction type from the transaction table joined to the surveys should get you what you want.   There is also the ma-trans table if you are looking at it via mailings.  

    • Jess Campbell

      Hi,

      Our integrator created a custom field for us called survey sent. When the survey is triggered by business rules, the incident custom field survey sent is set to yes. This is a super easy way to identify for which incidents a survey was sent. (Easier than combing through transactions for each incident.)

      If you want to look for the info by transaction, the table association will be incident, the transaction type will be 24 (Sent), you can transform attribute1 into readable data by using the following formula: if(transactions.id1>0, msg_lookup(31927)), which should populate the column with Survey Sent. Description will show the email address to which the survey was sent. You will want to severely limit the filters to pull only this one transaction for any incident which you want to verify.  Transaction Type= 24, Transaction Table = Incidents, Transaction Attribute 1 >0, Incident ID in list (your list of email incidents for these two agents). I would choose a sample of them, as too many incidents combined with the transaction table being ridiculously large will make your report not run.

      You may also want to consider whether there are other conditions on the survey trigger (like opt-ins, or recency, or frequency) which are stopping some from being sent. Additionally, we have found that some groups of our customers are more likely than others to respond, so if the agents in question are more likely to handle a different product line or type of incident than other agents in the same profile, you may see a reduced response rate based on that as well.

      Just some things to consider. Good luck!

       

    • Danette Beal

      Great comment Jess! yesyes

    • Jess Campbell

      Thanks Danette! Someone should use all the survey knowledge I have accumulated!