Technology leader and co-founder of Opkey — a continuous testing platform redefining test automation for web, mobile and ERP applications.
Many business and technology leaders realize that their digital transformation initiatives can’t be utilized without modernizing their enterprise resource planning (ERP) software. Incorporating new technologies such as artificial intelligence and machine learning is essential to modernizing ERP solutions.
Through a 2019 study of ERP migration and transformation projects, McKinsey revealed that two-thirds of enterprises did not get the ROI they were looking for from their migration project. The common reasons for this dissatisfaction are delays in ERP implementations and misaligned project goals. Intelligent test automation, which powers a continuous testing approach, will help ERP transformation projects run on time and stay within budget.
Continuous testing for ERP applications: Why do you need it?
Next-gen ERPs and digital operations platforms require innovative software to be released rapidly, with minimal business risk. Leading analysts from Gartner, Forrester (paywall) and IDC (registration required) now recognize that software testing in its current form cannot handle the challenges posed by ERP applications. These analysts have concluded that software testing must be aligned with DevOps and AgileOps practices to handle giant ERP transformation projects.
The Agile/DevOps approach is incomplete, inefficient and ineffective without continuous testing. In ERP migration projects where platforms are extended to incorporate new features, functionalities and technologies, continuous testing helps you transparently validate the performance of critical business processes. This significantly reduces the risks associated with a new implementation, along with scheduled software updates. By catching bugs early in the development cycle, continuous testing ensures minimal time and budget overruns while providing advantages in risk reduction.
What are the testing challenges of ERP transformations?
According to a report by Bloor (registration required), more than 80% of migration projects ran over budget in 2007. While I have seen that statistic improve over the years, I know migration projects regularly face issues of running over budget and over time. A 2019 ERP report from Panorama Consulting Group (registration required) shows that 45% of respondents had an average budget overrun of 30%.
Here are some specific testing challenges.
• Unclear Testing Scope: Determining what to test remains a major challenge for QA teams. The business risk grows every time too little testing is done. If you test too much, it wastes the time and resources of your business users.
• Inadequate Test Coverage: There are many moving parts in any ERP migration project. Functional and nonfunctional attributes get added, updated or removed with these migrations. Testing needs to pass various stages, from a unit test to a volume test, and eventually a mock go-live cutover.
• Change Frequency: In a recent Deloitte CIO survey, almost 45% of respondents reported that managing changes in an ERP project scope is one of the top frustrations in planning their ERP journey (pg. 10).
• Testing Fatigue: ERP projects are long and tedious processes. Using a manual testing methodology for ERP transformations can be inefficient and error-prone. Ask yourself: “Can my business users give their full effort to testing?”
Continuous testing for ERP applications: How can I make it work?
To incorporate continuous testing for a digital transformation, leaders must utilize automation. Teams should now focus on next-generation automation platforms that allow them to quickly build test cases, automate them and build the infrastructure to run them in a continuous fashion. Let’s review the four pillars of a continuous testing strategy.
• Know your ideal coverage: Here are some questions to ask yourself: “What’s my current test coverage? Am I testing all of our critical processes? If something goes seriously wrong, is it because I didn’t test enough?”
If the test cases you are automating only cover 30% of your core business processes, the automation might not be good enough. Emphasize knowing your ideal coverage and leverage a process mining technology to validate your ideal coverage. Test mining techniques surface your existing test cases, business processes and configurations from your system process log to determine your existing testing baseline.
• Apply continuous test development: Test assets require considerable reworks to keep pace with the frequent ERP changes typical in an accelerated release cycle. This speed cannot be achieved with continuous testing.
• Monitor changes continually: Ask yourself: “What has changed in the most recent ERP quarterly update? What business processes or test cases are going to be impacted?”
Emphasize the importance of knowing whether you are testing what is needed. Before the updates are pushed to production, use automation tools that give better change visibility to users by alerting them of processes that will be impacted.
• Test execution at scale: Prepare a scalable infrastructure to run thousands of tests on-demand with every change. Opt for a platform that can run your tests continuously on-premises, in the cloud and on mobile seamlessly.
What do you need from a test automation tool?
Three key capabilities must exist in a test automation tool to support an ERP transformation’s continuous testing paradigm.
• Autonomous Configuration Of Tests: Many changes happen at the configuration level for any ERP transformation. Leaders should leverage an automation tool that can autonomously create relevant data sets for test execution.
• Continual Impact Analysis: In the ERP world, updates are rolled out frequently. QA teams can find it difficult to decide the minimum number of test cases that need to be executed to ensure business continuity in post-application updates. AI-based impact analysis recommends a minimum number of test cases that need to be executed based on highlighted risks, keeping business application disruptions at bay.
• Autonomous Self-Healing Tests: QA teams often struggle to continuously maintain test scripts with each new release. Through leveraging AI-powered self-healing capabilities, changes can be identified automatically and test scripts can be fixed autonomously.
Continuous Test Automation: A Summary
The key to successful AgileOps is releasing updates as early and as often as possible.
With enterprise application vendors like Oracle, Microsoft and SAP rolling out updates on a weekly, monthly or quarterly basis, enterprises need to embrace those updates as early as possible. However, supporting your software testing initiatives will only be achieved with the right continuous testing strategy.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?