BizDevOps in a nutshell
Many organisations claim to be data-centric and data-driven. Making informed decisions based on what their organisation’s data is telling them. But how many organisations can truly say they have an effective and operational Data Management framework in place? The good news is that it is relatively easy to remove the toil and burden with automation and implementing a Data Model program to improve the accuracy and quality of the underlying data. But as always, there are wider benefits that make this even more important for organisations to build into their data management process.
Apart from setting a solid foundation in a Data Governance program, there’s a neat benefit for the business too. That is building flexibility, or future proofing, a mission critical or business critical application. Over time, the needs of the business change, business processes change, they evolve. Sometimes, even ‘The Business’ may not fully appreciate the full set of requirements at the point of inception. That’s where an Agile software delivery system has had a major impact.
In the world of DevOps and Agile Software Delivery, the database is the most rigid and inflexible part of the application and the one technology that needs the most future proofing. This rigidity is a major factor for applications not evolving over time as the business requirements change.
With this lens, implementing a Data Model program for application development is even more important for maximising business outcomes through the life of an application, whether be it mission critical, business critical, or otherwise. What’s more, the typical ‘cost of change’ after release is typically 100x or 1000x the cost of a change in design.
DevOps has had a tremendous impact as a process model, managing to tear down the walls between the development and operations teams. But there’s a growing need to build on the process model and technology has caught up enough to facilitate that. Enter DevOps 2.0 or BizDevOps – employing a Data Model is the key to improved data quality, productivity, controlled change and automation of database updates and deployments.
In a nutshell, the addition of “Biz” to DevOps acknowledges the input of business teams and their understanding of what the application should do to meet stakeholders’ needs. This important addition helps ensure that the organisation is solving actual problems, not creating software that does not align with business processes – or worse, creating solutions for problems that don’t exist.
The net result is an easy to modify data model, resulting in an easy to modify database, resulting in a much better application and business outcome. After becoming more familiar with the concept of BizDevOps, it is important to identify the tools and processes that will ultimately ensure a successful implementation.
Process 1: Business Process Modelling
One of the biggest challenges in undertaking the shift to BizDevOps is finding common terminology and understanding among business stakeholders and developers. Therefore, business process modelling can bridge this divide and make a real change.
In the context of BizDevOps, a business process modelling tool details the tasks, responsible parties, information elements involved in processes, and the interactions that occur across systems, procedures, and organisational hierarchies.
Furthermore, the business process modelling links business strategy to IT systems development and ensures business value by combining workflow, functional, organisational and data/resource views with underlying metrics such as costs, cycle times and responsibilities. As a result, both teams have on their hands a foundation for analysing value chains, activity-based costs, bottlenecks, critical paths and inefficiencies.
Process 2: Data Modelling
They are often overlooked after the initial ‘one and done’. But organisation’s data models are the key to a multitude of operational efficiency gains, quality improvement programs and business intelligence effectiveness.
When the DevOps teams can interoperate and integrate with the data model, this is where organisations can benefit from two of the core concepts of DevOps; collaboration and automation. The Data Model can capture changes implemented by the database development team, just as easily as the database development team captures the output of the Data Model. Thus, by utilizing a Data Model from the start, organisations are not only benefiting the business by capturing changes they need, but organisations are also improving quality and increase the speed of change from requirements to delivery. They are future proofing the database tier. They are ensuring flexibility and a capability to easily deliver change when needed.
Process 3: Metadata management
Another part of forming the common language among BizDevOps participants revolves around metadata management or the information the enterprise has about its data elements. Confusion reigns if there are different assumptions and conflicting definitions of data elements.
Metadata management tools help harvest, store and administrate the data that describes other data. These tools also track how data elements are related and their lineage. By centralising metadata in a data catalogue, both business and technical users can see how data flows through their business processes and can refer to that data with consistency and confidence. This benefit can’t be over emphasised. Just ask engineering teams charged to upgrade and merge multiple siloed applications into a modern central solution. The biggest challenge is always to get everyone speaking the same language and translating local to global.
Process 4. CI/CD tools
The success of BizDevOps also depends upon continuous integration and delivery of CI/CD tools, as they automate the workflow of the entire software development and delivery process, including building and testing the software, preparing the code base, and deploying code to destinations like containers, virtual machines and cloud servers.
The automation and the ongoing defect analysis results in errors reduction and enforces best practices for higher-quality code, ensuring that the software stays aligned with business requirements and can adjust quickly if those requirements change. This can be achieved through a shared metadata repository that enables the capture of changes from the business or IT and is presented to all stakeholders in a form they each understand.
Process 5. Collaboration
As discussed above, collaboration lies at the heart of successful BizDevOps, and many organisations need to make a conscious effort to include business stakeholders in the process, however, deploying the right tools can ensure software delivery meets business goals.
Conclusion
Overall, the success of BizDevOps relies on real-time analytics that will provide immediate insight into the metrics around application speed, load time, query response time and more. The continuous feedback loop, via real-time monitoring, enables teams to make smaller and more manageable adjustments to optimise the application and maximise the customer experience.
BizDevOps makes sure that communication breakdowns, lack of visibility into where data resides and a disconnected approach to safeguarding sensitive data do not occur, minimising the risk of a breach or non-compliance. However, it is important to keep in mind that to achieve success and speed up the creation of software, organisations need to break down the barriers between the business, IT, and development and stand-up a Data Modelling practice.
