Skip to content

Technical challenges when scaling process automation

Reading Time: 3 minutes

Automation of business processes is an opportunity for radical business transformation, it can improve operations and create better roles, with the potential of improving capacity and performance. By augmenting or replacing employee effort with automation’s that remove or reduce low value and/or repetitive work, businesses can lower operating costs and boost accuracy, speed, consistency and security. Successfully implementing and creating longevity, however, is complex, so here are my top five challenges:

1. Process pipeline

The goal is a healthy pipeline of automation opportunities, with a blend of processes from around the business that then gives flexibility and agility to the automation delivery team, who then serve this pipeline with an efficient SDLC. There are two critical factors:

  • The business is geared through a federated model to identify and assess processes for automation. This is required as centralising the automation analyst function is not scalable and does not provide the assessment speed or deep process detail
  • The business is incentivised to identify automation opportunities because of the drive to create better jobs and reduce costs, as well as the additional benefits of speed, accuracy, consistency and security

2. Infrastructure

Automation is usually introduced to or by divisions in a business seeking to transform their business area, therefore early engagement with IT (and subsequently IT requirements) are neglected, which can mean problems and delays when the automation programme commences. Automation tools are flexible and can be deployed on-premise or to the cloud. On-premise deployments are potentially more susceptible to infrastructure issues, such as network outages, certification issues, virus scanning interruptions, application/server patching cycles, group policy inconsistencies etc. Deploying in the cloud could mitigate a lot of these issues, with the benefits of the cloud aligning well with the automation of work (e.g. scalable servers/virtual workforce) and allowing your business to focus on its strengths which are possibly not IT infrastructure. Future considerations regarding infrastructure are likely to focus on how it can align with your delivery life cycle and support continuous deployment.

3. Team Priorities

For newly formed teams there will a host of demands; development & delivery, operational support, architectural design/evolution, infrastructure maintenance, data privacy & information security, reporting, programme governance. The rigours of these are likely to impact on your ability to scale to meet the demand/ambitions and may affect your attempts to optimise the flow of processes through the automation delivery lifecycle. DevOps can be a solution for newly formed automation teams as there are some key DevOps principles that inform team practices (that in turn resolve the challenges highlighted) principles such as; iterative, incremental, continuous, automated, self-service, collaborative, and holistic, with the complementary practices of self-service configuration, automated provisioning, continuous build, continuous integration, continuous delivery, automated release management, incremental testing. So understanding your team priorities and establishing a way of working (potentially based on established practices) is likely to assist your team mitigate or remove these challenges.

4. Change Management

Automating business processes will require business team/departmental change and depending on the scope may also have wider organisational change implications, because of the need for restructuring, process changes and cultural transitions. Change management should consider:

  • People change; by socialising the goals of automation and sharing the automation strategy
  • Automation benefits; ensuring success and confidence in the automation function and it should synchronize the required organisational change within the agreed timescales.  
  • Ways of working; helping evolve the business teams. A good example of this will be revising Contingency Planning which is vital for automation to integrate and continue to be an effective tool for the business

5. Testing and Releasing code

At the time of writing the Automation applications available on the market do not provide integration with popular and/or cloud-hosted testing or release tools and so at this point of the automation SDLC, testing code and releasing code to production can be time-consuming and so there is an inherent risk to the code quality. Therefore strong governance around the code base must be established and shared with all members of the automation, especially with contractors and third parties. This governance could be in the form of a Good Practice Coding Guide, a testing strategy supported by regular code reviews to ensure quality assurance. If this governance is not established there is a risk of a security breach and potentially reputational damage and government fines. The current challenge will be working with these automation tools and their limited testing and releasing functionality in an environment of agile delivery and continuous deployment/iteration. 

Categories

RPA, Scaling

Leave a Reply

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  Subscribe  
Notify of
%d bloggers like this: