Shell's finance organisation has largely centralised operational processes over past years into Finance & Data Operations (FO), which is located in four cost-advantaged locations. FO is driving a 'smart automation' strategy using low cost tools such as Robotic Process Automation (RPA) to make operational processes more efficient and effective. The goal is to free employees from repetitive, manual work and to focus their time on more value-adding activities. FO sees RPA primarily as a tool to reduce costs, but also recognises the additional benefits of improved process and data quality as well as having the flexibility to scale up during periods of peak demand.

For FO, RPA software is seen as 'one tool in the automation kitbag', and they see it as part of a broader continuous improvement strategy. This strategy is based on the ESSA approach - first try to eliminate waste in the process, if you can't eliminate a step, simplify it. Once it is simplified, look to standardise the processes across locations. Once these are standardised as much as possible, then look for the 'automation' opportunity around specific tasks or workflow, which may be RPA, may be a different automation tool, or may be RPA working in combination with other automation options. Without simplification and standardisation of finance processes, it is difficult to drive replication opportunities, it takes more time and is more difficult to code and build the robots, and they are not scalable. And RPA is not the solution to a badly designed process.

The combination of heavily streamlined processes with RPA technologies working together with other automation workflow tools and, (looking forward) more advanced machine learning capabilities to make a significant difference on return. Here FO sees the role of the technology solutions architect is critical in supporting the goals of the finance team, helping identify the most relevant automation technology or technologies to apply to the problem once the process maps have been identified, as well as supporting the programming, testing of software and ensuring appropriate integration of the software across the different applications. The selection of the right automation tools is critical, but it also requires individuals with the right skills to programme these emerging workflow automation tools.

An example of a typical project: after implementing ESS stages to the end to end process for updating customer master data fields, an RPA 'bot' was built to load the data requests in SAP. This was previously a manual and time consuming activity with lots handoffs and manual data validation steps. The new robot is used to update various kinds of SAP master data information in an automated manner and provides a 24x6 execution capability thereby increasing the bandwidth and responsiveness to the business. The average time taken to complete a request has reduced by 40% with 100% data quality being sustained month on month.