Within the Analysis module of MyFeelBack, indicators can be created from:
- from the library: Creating an indicator from the indicator library makes it available for several uses (dashboards, reports, alerts, etc.)
- from a dashboard: An indicator is displayed as a “thumbnail” within the dashboard. If an indicator is intended only for a specific dashboard, you can create the indicator directly from the dashboard. You can add it to your library afterwards, if necessary:
- from a report: the indicator appears within a page. You should select the page format before creating the desired indicator or select an existing indicator from your library.
What data should be used to form an indicator?
1. With attribute data
All the attribute data linked to the recorded answers may be used as a dimension or as a filter to form an indicator.
For example, you can use your products, sellers, categories, personas, countries attributes, etc.
If you're not sure you understand the concept of attribute, read our dedicated article!
2. With the collected data
3. With context data
The media (sms, email, website ...) and the platform (tablet, mobile, desktop) used.
4. With existing statistical data
We use an existing measurement as cross-linked data for comparison. For example, you can compare an overall NPS score with an NPS score by seller.
The 4 key terms to remember
This is the first level of selection when creating a new indicator. You should choose the type of data for your indicator to be based on:
- “Answers”: this is all the data recorded by answering to the surveys;
- “Visits” : it’s all visits within a campaign;
- “Solicitations”: is the number of times a respondent was asked to participate in the survey.
A dimension corresponds to the data that you will choose to display or to cross-link. The rendering of the indicator represents the crossing of the different dimensions, which you can choose to filter. Depending on the selection order or the data type, the dimensions will appear on the abscissa (horizontally, X axis) or on the ordinate (vertically, Y axis).
The filter will allow you to focus the analysis on a restricted view. For example, a specific time period, or a particular geographic area. Any available data can be used as a filter within an indicator.
The statistics will allow you to display a calculated value (average, variance, standard deviation, etc.) to have more consistent view. For example, displaying an average NPS by country would be more readable than a detailed view with a specific score per answer.
Indicator restitution goals
1. To show the distribution
Here is an example of restitution that helps you analyse the distribution of NPS score by seller:
2. To point out the evolution
In this example a line graph shows the evolution of the NPS score over time, by call reason, on a monthly scale. On this same graph with the main purpose is the evolution tracking, we will find interesting indicators of comparison: essential to situate performance over time, and also an induced notion of variation.
3. To illustrate the comparison
In this example, a "radar" view would be perfect for comparing the levels of importance of items by respondent category.
We can also consider some level of interest ranking by type of customer profile or some type of distribution, but these last points would't be the main goal of this indicator.
4. To indicate the value
A restitution of value allows to restore an average for a defined period, and thus to have a consolidated view with a control goal. It may be interesting to combine this information, like in this example of a variation in comparison with the previous time period.
For example, an overall satisfaction score out of 5:
Or to quickly complete with a detailed view on a distribution format.
For example, the distribution of an overall satisfaction level out of 5:
Summary of the different types of indicators and their graphical representations:
Now that indicators have no secrets for you, find out how to create your indicators step by step!