What technical and financial analytics should CIOs and decision makers expect from Enterprise Architects in 2022?
Enterprises are in the middle of an application explosion and a transformation acceleration.
Looking just at the application landscape: industry surveys tell us that the average enterprise is using 1,295 cloud services, and also runs around 500 custom applications. The worldwide enterprise applications market reached $241 billion last year, growing 4.1% year-over-year in 2020, according to IDC.
The underpinning architectures of enterprises– made up of interactions between people, processes and technology, and often also physical assets (IoT) – are also growing and changing at pace.
Enterprise Architects keep CIOs and business units informed using IT cost calculations and technical and lifecycle metrics.
They will often present costs and technical metrics for the current IT landscape, plus forecasts to inform planning for new business scenarios and digital transformation projects.
Common analysis in past years might have covered:
◉ Counts of applications
◉ Total cost of ownership of applications (also ROI and NPV)
◉ Which processes rely on a particular software or infrastructure (dependency analysis)
◉ How long technology is going to be supported, and when the enterprise needs to transition or upgrade
This basic data is useful but might still leave decision makers wanting additional analysis or tighter granularity.
Business units want to understand how much updated processes or applications will lead to improved technical metrics (uptime, responsiveness) or improvements to processes which are important for successful customer journeys.
Enterprise Architecture Data Analysis in 2022
Data-driven enterprise architecture can now provide greater detail and certainty around forecasts. Architects and business users need to design calculation which roll up data numerically across the architecture, generating required KPIs on-demand.
For the data scientists and numerate business analysts, steps such as Add, Subtract, Multiply, Divide, Min, Max, Average and Count are standard. In addition, operations such as Power, Log and Atan can be used to calculate trends, probabilities, attribute values and measure or predict impacts of business decisions.
As well as diagrams and roadmaps, EAs often need to be ready to provide reporting dashboards which include:
Technology Cost Analysis:
◉ On-demand totals of how much out-of-date technologies are costing the business
◉ Total cost of a specific Business Capability or Process – using the connections and relationships in the architecture to attribute portions of costs accurately
◉ More precise cost-of-ownership (e.g., calculating software, support or external services costs according to business function) Costs of underlying infrastructure or resources used. EAs can calculate the total cost of ownership of out-of-date technologies (available as a pre-built cost simulation in tools such as ABACUS from Avolution)
◉ Cloud migration costs
◉ Technical debt metrics such as remediation cost, complexity, cost of compliance
Lifecycles & Trends in Metrics
◉ Cost of risks and vulnerabilities associated to applications and technologies
◉ Technology and vendor lifecycle information summaries e.g., Number of years to retirement of a technology
◉ Application portfolio assessments: calculate and chart business fit and technical fit of applications and technologies. E.g., Are our applications using approved technologies, and are these technologies currently being supported by the vendor? (The base data for this analysis can be pulled in from sources such as Technopedia)
◉ Technical KPIs including Response Time, Availability, Reliability, Resource Utilizations
◉ Trends in metrics such as rate of growth in costs, or rate of increase in Availability or Reliability
◉ Machine-learning based predictions: E.g., Use lists of applications, lifecycles, financial data and other architectural content. For instance, the machine learning engine in ABACUS from Avolution provides a quantitative prediction of the values which belong in any empty cells. An ‘empty cell’ for an application, machine learning will propose a TIME (Tolerate-Invest-Migrate-Eliminate) recommendation, which the architects can choose to accept, for a more complete dataset.
Adding KPIs to Diagram-based Enterprise Architecture
◉ Comparison of future vs current state. Architects can dashboard side-by-side comparisons of information or technical architecture designs plus related catalogs and use these and related metrics to monitor transformation projects
◉ Risks associated with specific processes (security ratings and risk ratings can be rolled up from technologies and applications to the processes they support)
◉ Comparing tasks by mapping tasks or processes to capabilities. For instance, as part of a consolidation during a merger or acquisition, architects can calculate costs or technical KPIs on processes to determine efficiency of the two versions of the process.
◉ Dependency analysis of a process: using diagrams and Graph Views to see connections e.g. to highlight where a process is dependent on outdated technology.
◉ Show systems, interfaces and APIs as part of process diagrams
Analysis by Architects on data from APIs
Architects can also present data pulled in from CMBDs or other company data sources (via API queries) as stakeholder dashboards. Charts and interactive visuals are often clearer and easier to explain than lists of data.
◉ Architects can use tools such as Postman to run queries on APIs
◉ Common API integrations include Technopedia a range of CMDBs, and VMware products
We’re in the foothills of a golden age of architectural analysis. Businesses understand their external environment and run their sales functions with data from business intelligence tools and modern sales platforms. They are applying the same quantitative approach to their internal enterprises, a joined-up universe of data on people, processes and technologies.
Source: opengroup.org
Posts
0 comments:
Post a Comment