| Info |
|---|
Since version 3.1.0-KickOff, Ontologies have been referred to as "Entities" in the Control Panel. This does not alter any functionality; the nomenclature has simply been changed for a better understanding of the concept. |
| Table of Contents | ||
|---|---|---|
|
Updating this article
Introduction
A KPI (Key Performance Indicator) is mainly a metric to help us measure and quantify a progress following some started goals and objectives. That is to say, a KPI is designed to show how the progress is in a specific process or product.
KPIs are widely used in the world of marketing (where KPI are specific metrics (quantitative and qualitative) that allow you analyze each of the dissemination, performance, loyalty, positioning and branding channels in your publicity campaigns), and in general if I intend that my solution's objectives can be measured and reached, in other words, that they be SMART:
S for specific.
M for Measurable.
A for Attainable.
R for Realist.
T for Timely.
Steps
Create the Entity
From the Control Panel, navigate to the Development > My Entities menu.
...
This will take us to the list of available Entities. To create the Entity, click on the "+" button at the top right of the screen.
...
From the different Entities that we will be able to create, we will select the "Creation of a KPI Entity" one:
...
The platform allows you to generate KPIs easily. Within the platform's data-centric approach, the KPIs are ontologies and are generated by aggregating information that is stored in other ontologies.
Let's see an example showing how to create a KPI.
From the menu option DEVELOPMENT > My Ontologies, select CREATE:
...
In the next screen, select Creation KPI:
The creation screen asks for a Query. This input will be used to specify a query that will extract, group, transform and aggregate data from the ontology that serve as the basis for your KPI's creation. This query's result will be stored in the ontology that you will later select.
...
For this example, select the ontology Restaurants, including information on New York restaurants, their type of cuisine, position, etc.
...
Say you are interested in the rating of the restaurants by type of cuisine and borough, so your query will be:
select
...
c.Restaurant.cuisine
...
as
...
cuisine
...
,
...
c.Restaurant.borough
...
as
...
borough,sum(c.Restaurant.grades[0].score)
...
as
...
score
...
from
...
Restaurants
...
as
...
c
...
group
...
by
...
c.Restaurant.cuisine,c.Restaurant.borough
If you select it, you have:
...
Once you have the query (that allows you to use the NOW() function to establish time relationships such as, calculate last month's ratings),
use CRON's wiza4rd to assign an execution periodicity. In this case, say every day at 00:00:
...
You can also choose whether this KPI's generation will be made only between two dates or not (for instance, if you are working on a campaign and you want the KPI to be generated only during some specific days).
...
Finally, select whether you want to generate the KPI on a new ontology or on a pre-existing ontology (in case you generate several KPIs of different type on the same ontology):
...
In this case, select a new ontology and fill in the data:
...
After selecting Create, you can go back to the Ontology list, where you will see:
...
Here you can stop the task's execution:
...
Once the KPI has been created and the task is running, the execution is launched depending on the assigned temporality, and the query's resulting data are stored.