If the pandemic has taught organizations anything, it has been to quickly pivot to the changing demands of customers, opportunities and regulations. High-velocity IT allows for significant business enablement where time to market and time to change is critical. IT and business used to be wholly separate entities and behaved as such. With the advancement of digital technologies, IT is getting increasingly decentralized, being embedded into the lines of business by way of RPA. Automation has emerged as a unifying channel that can be used effectively to fast-track the product life cycle.
Change is the firing shot, and culture is the hurdle.
Our mobile apps ping us so many times a day. This is the soundtrack of change, one in which a minimum viable product is launched followed by subsequent incremental updates — some even daily. While the DevOps groups behind these products sit firmly in the agile driver’s seat, the ITOps and IT support areas remain set in their old waterfall ways.
If you are a developer, how often have you had to wait for an urgent software install thanks to the lengthy queue requests ahead of you? Why does one have to wait for approval? Why can’t employees be trusted to make legitimate requests? This is the new culture that high-velocity IT organizations need to lean into.
DevOps and testing are not in touch.
In a typical environment, a developer along with a few colleagues push code to the configuration manager. After the code is validated and merged in the core code, it is dispatched to testing. Bugs discovered are registered in a separate Jira-like system. Each bug is then sent back to the DevOps team to be fixed and closed in a GitHub-like development environment.
DevOps is looking for a responsive partner in IT(SM).
Raise your hands if you know of a DevOps team waiting for their Dockers and Kubernetes container requests to be fulfilled or twiddling their thumbs before their virtual machines arrive. Software installations and access provisioning should be the first port of call for automation. Even after software goes to production, linking to ticketing systems and deploying diagnostic script bots to achieve an end-to-end incident resolution are just some of the ways DevOps teams can self-serve.
State-of-the-art ML-based systems can automatically identify hotspots of high-volume incidents in an enterprise’s particular environment. Through the automation of service management using a combination of chatbots, VSAs, email bots and auto-ticket classifications, the entire cycle from ideation to production can be fast-tracked to high velocity.
A few best practices to achieve holistic high-velocity IT.
• Treating developers as customers. Providing an operations service catalog that lists the operational support services to support the application solutions for the business can go a long way toward simplifying complexity for developers subscribing to services.
• Simplifying change management. Standardizing and automating as many changes as possible would greatly simplify change management. Routine software updates can be standardized, and risk-prone areas can be provided with detailed risk analysis. Low-risk areas could go for a straight-through implementation and be made auditable, allowing for the reversible automation of implementation tasks.
• End-to-end automation should be the preferred goal. To eliminate errors and introduce agility in delivery, zero-touch automation with minimum human intervention should be advocated wherever possible. For manual steps like shipping hardware devices to users, partial automation will be unavoidable. Automated incident diagnosis remains a powerful way of reducing mean time to recovery. Based on our experience, we’ve found that automated monitoring and event response can allow for up to a 50% reduction of the time spent by NOC/SOC teams. Auto-filling requests before approval — and the subsequent retraction if not approved — can go a long way toward speeding up the fulfillment of requests.
• Centralized knowledge. Omnichannel access can facilitate faster learning and better decisions as a team. Consider incorporating technologies that make it easy to create, share and collaborate, as they could improve collective outcomes. AI/ML/process mining-based analytics can help detect process bottlenecks and identify waste. This can also be a preferred method to create high-value, incremental improvement initiatives.
Not everyone needs to be on high-velocity IT.
While high-velocity IT initiatives currently focus on speeding up product life cycles, those that affect customers and partners are not far behind. Systems in the enterprise that house systems of record change less frequently. Modules such as finance, procurement or production do not need to be part of high-velocity IT initiatives for now.
A time of reckoning for IT.
While software delivery has adopted an agile mindset, the operations side has not. What is the point of developing applications quickly and efficiently if software delivery is not equally quick? We are only as good as the weakest link. A central integration platform with an architecture that comes with ready plug-ins to the ever-growing landscape of tools across the enterprise is necessary for the next-gen enterprise.
From delivering digital engagement applications to providing stable operations with 24/7 availability, IT will need to step up its game. Game on.
By Uday Birajdar
CEO and co-founder at AutomationEdge
This article first published on Forbes here.