Analyzing agent data and automatically delivering actions

1. A server comprising:

one or more processors; and

one or more non-transitory computer-readable storage media to store instructions executable by the one or more processors to perform operations comprising:

receiving metrics data associated with a plurality of employees and gathered during a first time period, the metrics data received from a plurality of computing devices, wherein individual employees of the plurality of employees are associated with individual computing devices of the plurality of computing devices;

storing the metrics data in a database;

performing an analysis of at least a portion the metrics data;

generating, based on the analysis, one or more distributions of the plurality of employees based on key performance indicators in the metrics data;

generating, based on the one or more distributions, one or more persona models created by selecting metrics associated with an upper percentile of the one or more distributions of the plurality of employees;

determining to improve a performance of a set of employees with respect to at least one metric, based at least in part on comparing the metrics data associated with the set of employee with the one or more persona models;

training a set of machine learning models comprising two or more machine learning models, based on the one or more persona models to determine one or more result effective variables associated with the key performance indicators;

generating, by a voting classifier, a weighted result based on a top ranked subset of results from the two or more machine learning models;

determining, based on the at least one metric and the weighted result, an initial training to provide to each employee in the set of employees;

after providing each employee in the set of employees with the initial training, receiving, from the plurality of computing devices, additional metrics gathered in a second time period subsequent to the first time period;

retraining at least one machine learning model in the set of machine learning models based on the additional metrics data;

performing a comparison of the additional metrics data relative to the one or more persona models based at least in part on the retraining;

determining, by the set of machine learning models and based on the comparison, a recommendation to improve a performance of a particular at least one employee of the plurality of employees;

selecting, based on a personality type associated with the at least one employee and based on the recommendation, an additional training; and

automatically delivering the additional training to at least one computing device associated with the at least one employee.

2. The server of claim 1 , wherein training the set of machine learning models based on the one or more persona models comprises determining one or more high performers based on the one or more persona models.

3. The server of claim 1 , wherein the metrics data comprises data indicating individual employees interacting with an associated computing device of the plurality of computing devices in a particular time period.

4. The server of claim 1 , wherein:

the two or more machine learning models comprise at least two of an extra trees classifier, a catboost classifier, an extreme gradient boost classifier, a light gradient boost machine, a gradient boosting classifier, quadratic discriminant analysis logic, linear discriminant analysis logic, a random forest classifier, logistic regression logic, naïve bayes, an ada boost classifier, a K-neighbors classifier, a decision tree classifier, a state vector machine (SVM) with a linear kernel, a ridge classifier, long short term memory logic, natural language processing logic, recurrent neural networks (RNN), convolutional neural networks (CNN), multi-level perceptrons, feedforward neural networks, or a combination thereof.

5. The server of claim 1 , the operations further comprising:

determining, based on the at least one metric, a type of the initial training to provide to individual employees in the set of employees;

determining a personality type associated with a particular employee in the set of employees;

selecting, based on the personality type associated with the particular employee and based on the type of training, a particular training module; and

delivering the particular training module to the particular employee.

6. The server of claim 5 , the operations further comprising:

determining, based on monitoring data from a particular computing device associated with the particular employee, a time period when the particular employee is available; and

delivering the particular training module to the particular employee during the time period when the particular employee is available.

7. The server of claim 5 , the operations further comprising:

determining that the particular employee has completed the particular training module;

gathering second data associated with the particular employee in the second time period; and

determining that the particular employee has improved with respect to:

the at least one metric relative to a previously determined performance of the particular employee; or

the at least one metric relative to other peers in a peer group that includes the particular employee.

8. The server of claim 5 , the operations further comprising:

determining that the particular employee has completed the particular training module;

gathering second data associated with the particular employee in a second time period that is subsequent to the particular employee completing the particular training module;

determining that the particular employee has not improved with respect to:

the at least one metric relative to a previously determined performance of the particular employee; and

the at least one metric relative to other peers in a peer group that includes the particular employee;

selecting, based on the personality type and on the type of training, a second particular training module; and

delivering the second particular training module to the particular employee.

9. The server of claim 5 , wherein the set of employees is below a particular percentile in a standard distribution of employees.

10. A method, comprising:

receiving, by one or more processors, metrics data associated with a plurality of employees and gathered during a first time period, the metrics data received from a plurality of computing devices, wherein individual employees of the plurality of employees are associated with individual computing devices of the plurality of computing devices;

storing, by the one or more processors, the metrics data in a database;

performing an analysis of at least a portion of the metrics data;

generating, by the one or more processors and based on the analysis, one or more distributions of the plurality of employees based on key performance indicators in the metrics data;

generating, by the one or more processors and based on the one or more distributions, one or more persona models created by selecting metrics associated with an upper percentile of the one or more distributions of the plurality of employees;

determining to improve a performance of a set of employees with respect to at least one metric, based at least in part on comparing the metrics data gathered associated with the set of employee with the one or more persona models;

training, by the one or more processors, two or more machine learning models based on the one or more persona models to determine one or more result effective variables associated with the key performance indicators;

generating, by a voting classifier, a weighted result based on a top ranked subset of results from the at least two machine learning models;

determining, based on the at least one metric and the weighted result, an initial training to provide to each employee in the set of employees;

after providing each employee in the set of employees with the initial training, receiving, from the plurality of computing devices, by the one or more processors, additional metrics gathered in a second time period subsequent to the first time period;

retraining at least one machine learning model in the set of machine learning models based on the additional metrics data;

performing a comparison of the additional metrics data relative to the one or more persona models based at least in part on the retraining;

determining, by the one or more machine learning models and based on the comparison, a recommendation to improve a performance of at least one employee of the plurality of employees;

selecting, based on a personality type associated with the at least one employee and based on the recommendation, an additional training; and

automatically delivering, by the one or more processors, the additional training to at least one computing device associated with the at least one employee.

11. The method of claim 10 , wherein training the two or more machine learning models comprises training of one or more classifiers that have been trained to identify and predict high performers based on the one or more persona models.

12. The method of claim 10 , wherein:

the metrics data comprises data associated with individual employees interacting with an associated computing device of the plurality of computing devices received in the first time period; and

the additional training results in the at least one employee moving from a below-average portion of the one or more distributions to an above-average portion of the one or more distributions.

13. The method of claim 12 , further comprising:

determining a measured outcome based on the additional metrics data;

performing a comparison of the expected outcome to the measured outcome;

determining an additional recommendation based on the comparison; and

automatically delivering the additional recommendation to the particular computing device associated with the particular employee.

14. The method of claim 12 , wherein: the two or more machine learning models comprise at least two of an extra trees classifier, a catboost classifier, an extreme gradient boost classifier, a light gradient boost machine, a gradient boosting classifier, quadratic discriminant analysis logic, linear discriminant analysis logic, a random forest classifier, logistic regression logic, naïve bayes, an ada boost classifier, a K-neighbors classifier, a decision tree classifier, a state vector machine (SVM) with a linear kernel, a ridge classifier, long short term memory logic, natural language processing logic, recurrent neural networks (RNN), convolutional neural networks (CNN), multi-level perceptrons, feedforward neural networks, or a combination thereof.

15. The method of claim 10 , further comprising:

determining, based on the at least one metric, a type of the initial training to provide individual employees in the set of employees based on a variance of individual employees relative to a high performer persona of the one or more personal models;

determining a personality type associated with a particular employee in the set of employees;

selecting, based on the personality type and based on the type of training, a particular training module; and

delivering the particular training module to the particular employee.

16. The method of claim 15 , further comprising:

determining that the particular employee has completed the particular training module;

gathering second data associated with the particular employee in the second time period that is subsequent to the particular employee completing the particular training module; and

determining that the particular employee has improved with respect to:

the at least one metric relative to a previously determined performance of the particular employee; and

the at least one metric relative to other peers in a peer group that includes the particular employee.

17. One or more non-transitory computer-readable media storing instructions executable by one or more processors to perform operations comprising:

receiving metrics data associated with a plurality of employees and gathered during a first time period, the metrics data received from a plurality of computing devices, wherein individual employees of the plurality of employees are associated with individual computing devices of the plurality of computing devices;

storing the metrics data in a database, the database comprising non-transitory computer readable media storing the metrics data in computer readable form;

performing an analysis of at least a portion of the metrics data;

generating, based on the analysis, one or more distributions of the plurality of employees based on key performance indicators in the metrics data;

generating, based on the one or more distributions, one or more persona models created by selecting metrics associated with an upper percentile of the one or more distributions of the plurality of employees;

determining to improve a performance of a set of employees with respect to at least one metric, based at least in part on comparing the metrics data associated with the set of employee with the one or more persona models;

training two or more machine learning models based on the one or more persona models to determine one or more result effective variables associated with the key performance indicators;

generating, by a voting classifier, a weighted result based on a top ranked subset of results from the two or more machine learning models;

determining, based on the at least one metric and the weighted result an initial training to provide to each employee in the set of employees;

after providing each employee in the set of employees with the initial training, receiving, from the plurality of computing devices, additional metrics gathered in a second time period subsequent to the first time period;

retraining at least one machine learning model in the set of machine learning models based on the additional metrics data;

performing a comparison of the additional metrics data relative to the one or more persona models based at least in part on the retraining;

determining, by the one or more machine learning models and based on the comparison, a recommendation to improve a performance at least one employee of the plurality of employees;

selecting, based on a personality type associated with the at least one employee and based on the recommendation, an additional training; and

automatically delivering the additional training to at least one computing device associated with the at least one employee.

18. The one or more non-transitory computer-readable media of claim 17 , wherein:

the metrics data comprises data associated with individual employees interacting with an associated computing device of the plurality of computing devices received in the first time period; and

the recommendation has an expected outcome comprising moving the particular employee that is in a below-average portion of the one or more distributions to an above-average portion of the one or more distributions.

19. The one or more non-transitory computer-readable media of claim 18 , further comprising:

determining a measured outcome based on the additional metrics data;

performing a comparison of the expected outcome to the measured outcome;

determining an additional recommendation based on the comparison; and

automatically delivering the additional recommendation to the particular computing device associated with the particular employee.

20. The one or more non-transitory computer-readable media of claim 17 , further comprising:

determining, based on the at least one metric, a type of the initial training to provide to individual employees in the set of employees;

determining a personality type associated with a particular employee in the set of employees;

selecting, based on the personality type associated with the particular employee and based on the type of training, a particular training module; and

delivering the particular training module to the particular employee.