Reports API
Yva.ai Reports API allows you to access and download the resources about your company connected to Yva.ai.
Make sure to get the Reports API token as it is described here: https://findoteam.atlassian.net/wiki/spaces/YHCE/pages/1551925263/Yva.ai+API#Reports-API-token.
Call a Yva.ai API method
To get specific or general information you should choose a request from the methods listed in the technical documentation: https://api-docs.yva.ai/#tag/Reports .
Get information about groups
There are two methods to get information about groups in your team in Yva.ai API:
Both methods allow getting the general system information about groups such as:
Type of the group,
Name of the group,
Number of group members,
Group ID in the system.
Get attributes of users
There are two methods to get attributes of users in Yva.ai API:
Both methods allow getting attributes such as:
Employee name,
Employee email,
Employee location,
Employee date of birth,
Employee gender,
Employee job title,
Employee department,
Manager name of an employee,
Manager email of an employee,
Employee management level,
Employee potential level,
Employee performance level,
Employee hire date,
Employee termination date,
Employee employment type,
Employee remote work (yes or no),
Custom 1,
Custom 2,
Custom 3,
Employee ID.
Get burnout metrics of groups
There are two methods to get burnout metrics of groups in Yva.ai API:
To use these methods you will need to specify the start date (should be Sunday) and the end date (should be Saturday).
Both methods allow getting metrics for the each week of the period such as:
Passive Yva.ai health index of group,
Number and % of employees in each burnout stage.
Yva.ai defines 4 burnout stages:
Green or without burnout signs,
Yellow or with early burnout signs,
Red or with late burnout signs,
Not enough data.
Calculation of metrics
Passive Yva.ai health index of group
Passive Yva.ai health index of group is calculated as follows:
100*(% of employees without burnout signs - % of employees with late burnout signs + 100%)/2Index is calculated on the end-date of the last ended week of the period. Index does not include employees, for whom there is not enough data.
Get burnout metrics for users
There are two methods to get burnout metrics of groups in Yva.ai API:
To use these methods you will need to specify the start date (should be Sunday) and the end date (should be Saturday).
Both methods allow to get metrics for each week of the period such as:
Wellbeing/Burnout stage,
Number of weeks in the current stage stage,
Data sufficiency indicator,
Yva wellbeing/burnout index.
Get active metrics of groups
There are two methods to get active metrics of groups in Yva.ai API:
To use these methods you will need to specify the start date (should be Sunday) and the end date (should be Saturday). By default the period of request is equal to 13 weeks but it can be changed.
Both methods allow to get metrics for the whole period such as:
Number of employees with activated surveys,
Number of employees with activated gathering passive data,
The response rate,
The participation rate,
The completion rate,
% of positive, neutral and negative responses to questions.
In addition to that, the response will include the following metrics for each week:
Question metrics for each question group,
Active Yva.ai health index of group.
Question metrics for each question group include:
ID of question group,
Name of question group,
% of positive, neutral and negative responses to questions of question group,
Set of question metrics for each question.
Question metrics for each question include:
ID of question,
Number of answers,
Number of positive, neutral and negative questions,
% of positive, neutral and negative questions,
Number of employees answered on question,
Number of employees answered positively, neutrally and negatively on question,
Text of question.
Calculation of metrics
Response rate
Response rate of the week is calculated as follows:
(number of all answers (including skips) on engagement questions during the week)/
(number of given engagment questions during this week)Response rate of the period is calculated as follows:
(number of all answers (including skips) on engagement questions during the period)/
(number of given engagment questions during the period)Participation rate
Participation rate of the week is calculated as follows:
(number of unique employees that gave answer (including skips) on at least one engagement questions during the week)/
(number of unique employees that were given engagment questions during the week)Participation rate of the period is calculated as follows:
(number of unique employees that gave answer (including skips) on at least one engagement questions during the period)/
(number of unique employees that were given engagment questions during the period)
This metric is calculated on the basis of employee belonging to the observable group on the last date of the period.Completion rate
Completion rate of the week is calculated as follows:
(number of surveys wuth answers (including skips) for ALL engagement questions during the week)/
(number of surveys with engagement questions with at least one answer on engagement question (including skips) during the week)Completion rate of the period is calculated as follows:
(number of surveys wuth answers (including skips) for ALL engagement questions during the period)/
(number of surveys with engagement questions with at least one answer on engagement question (including skips) during the period)Active Yva.ai health index of group
Active Yva.ai health index of group is calculated as follows:
100%*((% of positive responses for the period - % of negative responses for the period)+100%)/2Index is calculate on the basis of all EX questions.
Yva.ai considers positive answers as the answers from 7 to 10.
Yva.ai considers negative answers as the answers from 0 to 3.
Get 360 metrics for users
There are two methods to get 360 metrics of users in Yva.ai API:
Both methods allow to get metrics for such as:
Social capital index,
Set of ranks for each style of leadership,
Employee ID.
Yva.ai identifies 5 types of social capital ratings:
Integration social capital rating,
Expertise social capital rating,
Domination social capital rating,
Innovation social capital rating,
Protection social capital rating.
Calculation
Social capital rating for each style of leadership
For each employee, Yva.ai takes the number of positive responses on the three social capital questions related to a particular social capital type. Each positive response is equal to one point.
System ranks the entire list of employees in the company by the number of points scored - from the maximum to the minimum. Each employee is assigned a percentile by position in the list, so that the employee with position #1 in the list has a 100 percentile, and the last one in the list has a 0 percentile. For a specific employee, Yva.ai outputs this value in %, i.e. the first person in the list will get the value 100%, and the last one will get 0%. Employees for whom there were no answers on three key social capital questions in the last X weeks receive the value of the metric "n/a". By default, metrics are calculated on the period X=13 weeks, but this period can be changed depending on the request in the API.
Overall social capital rating
For each employee, the arithmetic mean of Protection, Innovation, Domination, Expertise, and Integration capitals is calculated. The values of “n\a” are taken as zero for calculations (i.e. if an employee does not have some of the types of social capital, then this will reduce his/her overall social capital index). By default, the metric is calculated on the period X=13 weeks, but this period can be changed depending on the request in the API.
Get passive metrics for users
There are two methods to get passive metrics of users in Yva.ai API:
To use these methods you will need to specify the start date (should be Sunday) and the end date (should be Saturday).
Both methods allow to get metrics for the whole period and for each week such as:
Set of metrics of digital work day,
Digital activity,
Email metrics for each communication type and message class,
Set of other sources metrics,
Employee ID.
Metrics of digital work day
These metrics include:
Digital activity start time,
Digital activity end time,
Digital workday length,
Communication hours.
Calculation
Digital activity start time - shows the average time of the first communication of an employee per week (7 days). For large periods, you are shown the median of weekly values for the selected period. If an activity is performed in the interval from 12 pm to 4 am, then the system refers it to the previous day as the last one.
Digital activity end time - shows the average time of the last communication of an employee per week (7 days). For large periods, you are shown the median of weekly values for the selected period. If an activity is performed in the interval from 12 pm to 4 am, then the system refers it to the previous day as the last one.
Digital workday length - shows the average time between the first and the last communication of an employee per week (7 days). For large periods, you are shown the median of weekly values for the selected period.
Communication hours - It is counted as the sum of the intersecting time periods of employee activities per week. If you select a period more than a week in the report, then the value is averaged and you are shown the median of weekly values for the selected period. Moreover, the system has its own duration set for each type of activity.
Email metrics (communication type)
Email metrics are related to the general quantity of emails received/sent, percentage of responses and response time to emails.
These metrics can be divided into three communication types. First one describes messages from all contacts and to all contacts. Second one describes internal communications between employees and the third one - describes communications with external contacts (other domains).
Metric are:
Number of received emails - external, internal, all contacts
Number of sent emails - external, internal, all contacts
Average response time to emails - external, internal, all contacts
Share of responses to emails - external, internal, all contacts
Calculation
Number of received emails - shows amount of emails received by employee per week (7 days). For large periods, you are shown the sum of weekly values for the selected period. Automatic notifications and mailings are not included in calculations. This metric is calculated for messages between employee and all of his contacts (all contacts), for messages between the employee and all of his contacts within the company (internal contacts) and for messages between the employee and all of his contacts outside the company (external contacts).
Number of sent emails - shows amount of emails sent by employee per week (7 days). For large periods, you are shown the sum of weekly values for the selected period. Automatic notifications are not included in calculations. This metric is calculated for messages between employee and all of his contacts (all contacts), for messages between the employee and all of his contacts within the company (internal contacts) and for messages between the employee and all of his contacts outside the company (external contacts).
Average response time to emails - For a weekly calculation, the system uses all emails sent by the user in a week and checks whether a specific email was a response to a previous email. For each email that is a response, the system uses the difference in HH:MM:SS between the moment the reply was sent and the moment the question was received. The median of all obtained values is applied for a week. For large periods, you are shown the median of weekly values for the selected period. This metric is calculated for messages between employee and all of his contacts (all contacts), for messages between the employee and all of his contacts within the company (internal contacts) and for messages between the employee and all of his contacts outside the company (external contacts).
Share of responses to emails - For a weekly calculation, the system uses all emails sent by the user in a week and checks whether a specific email was a response to a previous email. Then the system calculates the share of all responses to emails. For large periods, you are shown the median of weekly values for the selected period. This metric is calculated for messages between employee and all of his contacts (all contacts), for messages between the employee and all of his contacts within the company (internal contacts) and for messages between the employee and all of his contacts outside the company (external contacts).
Email metrics (message class)
For each communication type there are 5 message classes. For each of them you will receive the following metrics:
Number of received emails,
Number of sent emails,
Average response time to emails,
Percentage of responded emails.
Message classes are:
Messages with tasks,
Messages with positive sentiment,
Messages with negative sentiment,
Messages with praise,
Messages with conflict.
Other sources metrics
These metrics are related to other activities of employee such as:
Number of sent chat messages,
Number of sent chat reactions,
Number of other activities,
Number of Jira Issue activities,
Number of Jira Comment activities,
Number of Jira Worklog activities,
Number of chat mentions.
Get ONA metrics
There is one method to get ONA metrics Yva.ai API:
To use this method you will need to specify the start date and the end date. Additionally, you will need to specify the minimal weight of edge and its balance.
Minimal weight of edge is the number of minimal messages between two employees to consider it as a well-established connection (edge). All communication pairs that are below this number will not be included in further calculation.
Edge balance is a parameter that will take away all the single-edged communications. We recommend using the 0.2 value of the parameter. All the edges below this value will not be included in further calculation.
This method allows to get metrics such as:
Directed edges of graph,
Node metrics,
Employee ID.
Directed edges of graph include:
Source of edge as the ID of employee that sends messages,
Target of edge as the ID of employee that receives messages,
Set of number of messages for each message class,
Messages with task,
Messages with positive sentiment,
Messages with positive sentiment,
Messages with praise,
Messages with conflict.
Weight of edge (total number of messages).
Node metrics include:
Degree,
Community (cluster),
Degree centrality,
Eigenvector centrality,
Closeness centrality,
Betweenness centrality.