Identifying contributing factors to a metric anomaly

A non-transitory computer-readable medium storing instructions that, when executed by at least one processor, cause a computing device to:

• receive, from a network server, web traffic data corresponding to a website, wherein the web traffic data comprises user action-based metric values indicating a particular type of user action performed by users in connection with a website or native application;
• store the web traffic data from the network server in a data storage associated with an analytics manager;
• analyze, utilizing a server of the analytics manager interacting with an analytics database, the web traffic data from the network server to identify at least one metric value anomaly indicating the particular type of user action;
• retrieve the web traffic data from the data storage;
• identify, utilizing a machine learning model that is trained on data associated with the at least one metric value anomaly from the analytics database to analyze the web traffic data retrieved from the data storage, a plurality of contributing factors that statistically contributed to the at least one metric value anomaly, the plurality of contributing factors comprising actions leading to the at least one metric value anomaly indicating the particular type of user action and the plurality of contributing factors are actions different than the at least one metric value anomaly indicating the particular type of user action;
• determine, utilizing the machine learning model trained on the data from the analytics database, a statistical significance for each of the plurality of contributing factors that contributed to the at least one metric value anomaly in order to identify correlations between the plurality of contributing factors and the at least one metric value anomaly;
• and
• present a subset of the plurality of contributing factors that have a highest statistical contribution to the at least one metric value anomaly.

A system comprising:

• an analytics database;
• and
• at least one processor configured to cause the system to:
• receive, from a network server, web traffic data corresponding to a website, wherein the web traffic data comprises user action-based metric values indicating a particular type of user action performed by users in connection with a website or native application;
• store the web traffic data from the network server in a data storage associated with an analytics manager;
• analyze, utilizing a server of the analytics manager interacting with an analytics database, the web traffic data from the network server;
• identify at least one metric value anomaly indicating the particular type of user action from the web traffic data comprising the user action-based metric values;
• retrieve the web traffic data from the data storage;
• identify, utilizing a machine learning model that is trained on data associated with the at least one metric value anomaly from the analytics database to analyze the web traffic data retrieved from the data storage, a plurality of contributing factors that statistically contributed to the at least one metric value anomaly, the plurality of contributing factors comprising actions leading to the at least one metric value anomaly indicating the particular type of user action and the plurality of contributing factors are actions different than the at least one metric value anomaly indicating the particular type of user action;
• determine, utilizing the machine learning model trained on the data from the analytics database, a statistical significance for each of the plurality of contributing factors that contributed to the at least one metric value anomaly in order to identify correlations between the plurality of contributing factors and the at least one metric value anomaly;
• and
• present, within a second portion of a display, a subset of the plurality of contributing factors that have a highest statistical contribution to the at least one metric value anomaly.