Scrutinizing the results of a campaign
Overview
After you publish a campaign, the targeted end-users start answering the questions that constitute the campaign. Learn here how to examine in detail the answers that users provide to your campaigns, either by creating investigations in the Finder or by building new dashboards in the Portal.
Quick campaign results
To quickly display the results of an ongoing campaign so far:
Log in to the Finder as a user with the right to create campaigns.
Select the Campaigns section on the left-hand side of the main window.
Right-click the name of a published campaign.
Select Display results from the menu.
You get a list of all the users along with their campaign-related fields. Refine the obtained results by adding or removing columns and conditions to this automatically generated investigation.
User fields related to campaigns
Once you create a campaign, a new set of custom fields related to the campaign are available in all objects of type user. These fields let you see the responses of every user to each one of the questions in the campaign, as well as additional information according to the status of the user in relation to the campaign. These fields are also available as conditions on user objects in investigations.
The names of all campaign-related fields start with the name of the campaign and have the following values:
Campaign name - Last action time
The last time that the user interacted with a campaign (dealt with a notification or answered any question).
Campaign name - Number of days since last action
How long since the user interacted with a campaign for the last time.
Campaign name - Status
How far the user has gone through the campaign.
Campaign name - Question name
There is one additional field per question defined in the campaign. The value of this field corresponds to the answer given by the user:
For single answer questions, the value is the name of the answer selected by the user.
For multiple answer questions, the value is a list with the names of all the answers selected by the user.
For opinion scale questions, the value is the name of the answer selected by the user, followed by the numerical value attributed to the answer between parentheses.
Campaign name - Question name (optional text)
For multiple answer or opinion scale questions that include an optional free text field, the value is the text typed in by the user.
Status of a user
The status of a user in relation to a campaign depends on whether the user accepts to participate in it or not and, in the former case, on how far the user went into answering the questions of the campaign. The status of the user regarding the campaign changes:
Unknown
The campaign did not target the end-user or the end-user did not receive the notification yet, either because of problems communicating with the user's device or because the device is turned off.
Notified
The user received the notification from the campaign but ignored it.
Opened
The user clicked on the notification to answer the campaign but did not answer any question.
Partially answered
The user started to answer the questions of the campaign but did not go to the end of it.
Fully answered
The user answered all the questions of the campaign.
Postponed
After receiving the notification, the user decided not to immediately take the survey but be reminded about it in the future (the user clicked the button Later).
Declined
After receiving the notification, the user decided not to take the survey (the user clicked the button Never).
Delayed (do not disturb)
The campaign has not been delivered yet because it arrived during an interval that protects the user from bursts of campaigns. The user will receive the campaign once the protection period elapses.
You can run investigations with conditions on users using the created campaign-related Custom Fields.
Metric computations related to campaigns
Campaigns also let you define metrics around them. Build dashboards in the Portal that display campaign-related metrics to analyze the responses of your end-users and measure the success of your campaigns.
Count metrics
Metrics that count the number of users may benefit from the campaign-related custom fields of users to group the results. When displaying the results of a count metric in the Portal, either in a table or in bar chart widget, you can group the users by their response to a particular question or by their status in connection with a campaign.
In this way, follow the progress of a campaign by creating a count metric that groups users by their status in relation to the campaign. Add the metric to a bar chart widget in a dashboard and you will get a breakdown with the number of users who fully answered the campaign, those who only answered a few questions, those who answered none, etc.
In addition to grouping results, campaign-related custom fields of users let you define ratios in count metrics. For instance, if you define a metric that gets the number of users who both fully answered the campaign and answered positively to a particular question and then compares it to the total number of users who fully answered the campaign, you get a count metric that computes the ratio of users who answered positively.
It is recommended that you always use the modifier all users and not active users when defining count metrics that relate to end-user feedback. The reason is that you usually want to take into account the responses of all users who answered the campaign, no matter if they were not active the last day.
Quantity metrics
Similarly to the generated custom fields for users that were mentioned above, creating a campaign also adds new options to create quantity metrics based on users. Since quantity metrics require a numerical value, only opinion scale questions make these options available. Thus, for every opinion scale question included in a campaign, there is a corresponding option to create a quantity metric for objects of type user. The label of the name has the following format:
Campaign name - Opinion scale question
Quantity metrics related to opinion scale questions make easy to compute average response values in the Portal.
RELATED REFERENCE
Last updated