Analytics in application delivery has greatly evolved in the last few years. We have seen clients moving from centers of excellence to agile organizations where business lines have the freedom to operate individually or centralizing delivery functions, either internally or outsourced. Sometimes both transitions happen within the same organization concurrently!
In all scenarios, however, there is a clear need for metrics, whether a centralized solution, providing ‘one version of the truth’; or a flexible solution providing a customized, self-service experience that is different across business lines. Complicating things further, is the need to combine information from multiple systems. This is especially true in modern DevOps environments. One recent example included Micro Focus ALM
, Silk Test, Jira, Jenkins and other tools.
An analytics platform in the current landscape with Agile, DevOps and digital transformations must address these 6 key areas:
Automatically gather information from multiple data sources in a meaningful manner
Making sense of data from multiple systems and connecting it together in a meaningful manner requires a data model where the entities and the relationships between them are well defined. The data model must be 'analytics friendly' and the way the data is collected and stored must provide historical value for trend and benchmark analysis.
Ability to deal with large volumes of data over time
Large volumes of data require special attention. Several areas should be addressed – the time it takes to load the data, the ability to recover from disruptions in the data load, minimal performance impact on the source system and that the data structure supports analysis on the relevant portion of the data required by the analytic use case.
Provide central dashboards (metrics & KPIs)
Central dashboards are crucial to obtaining one version of the truth; however, they do present challenges – the ability to provide value to groups that work differently or have different requirements, ability to adapt quickly to changes in organizational priorities, etc.
Provide ‘controlled’ self-service
Self-service is powerful in the hands of analysts but giving complete freedom will result in different groups reinventing the wheel and loss of standardization in the ways metrics are created and consumed. A good balance is achieved by using ‘controlled’ self-service – where the analytics building blocks are defined in the platform and the analysts use those pre-defined metrics, calculations, and visualizations to create their own analytics.
Include industry standard metric and benchmarks
Working many years with analytics in the app delivery domain, we have realized it is extremely challenging for organizations to correctly define what they need to measure. Even when they do have these definitions, it is challenging to agree what are the right thresholds and what to benchmark against.
Drive adoption by the user community
Adoption by the user community is achieved by creating excitement and engagement. This is achieved by addressing the community’s pressing problems, providing immediate value and of using modern and attractive UI.
To learn more about the process to integrate analytics into your environment, join us on September 13th at 11:00 am ET (8:00 am PT) for a VIVIT webinar