RPA developers have long relied on business users and analysts to understand processes and identify automation opportunities. However, analysis and digitisation tools like Conversational Data Intelligence, Process Mining and Intelligent Document Processing, are providing entirely new data sources to extend the scope and success of RPA.
Successful automation needs good data. RPA developers need access to clean, structured data to build their bots, improve performance and reduce exceptions over time.
But good data isn’t always easy to find. Sometimes, automation initiatives are doomed from the start - compromised by poor data and inaccurate insight from process discovery and definition.
Automating the wrong process won’t deliver the continuous improvement or seamless service that clients expect. You’ll also struggle to improve automations when you can’t measure their impact or see where they’re coming up short.
The challenge is that RPA devs often lack the full stack of tools needed for in-depth process analysis and data extraction. To deliver the most value from RPA, developers need supporting capabilities which discover, digitise and analyse processes from beginning to end.
Here are three of the most important tools for enabling RPA success and process efficiency:
Conversational Data Intelligence
Conversation and communications-based processes - everything from reading and responding to emails, to triaging customer queries - are one of the biggest barriers to automation.
The unstructured, natural language these processes are built on offer none of the structured, machine-readable data devs need to discover processes, implement automation or measure automation performance. As a result, they’ve become a no-man’s land for developers, and structured processes have typically been preferred for automation.
The problem is that this severely restricts the scope of automation success. Most business processes are mediated by communications at some point, making end-to-end automation impossible. Furthermore, email represents one of the most costly and inefficient workflows, costing businesses $5,000 to £10,000 a year per employee depending on company size.
RPA devs could realise considerable savings, if only they had the tools to process and automate communications en masse. Conversational Data Intelligence platforms now provide that capability. Combining state-of-the-art NLP with advanced analytics, these tools accurately convert natural language into structured data for analysis and automation.
Not only does this provide the structured data needed for building automations, it gives devs complete insight into business communications through a capability called Communications Mining. They can analyse the impact of their automations through the conversations of business users, while also surfacing new opportunities for automation and improvement. For the first time, low-level email and communications become fully understandable and automatable.
While Conversational Data Intelligence explains the ‘why’ of process breakdown, Process Mining defines the ‘how’. It focuses on the analysis of structured data - normally event logs - to give users a detailed and objective picture of process efficiency.
How long does a user spend thinking about their next action in a sequence, how many mouse clicks do they make each minute? Process Mining helps to answer these kinds of questions.
After a particular automation has been deployed, an RPA dev might leverage Process Mining to analyse and measure the benefits of the change programme. If the analysis shows little improvement, the dev can use Communications Mining to explore the reasons why, before making the necessary changes. A virtuous circle of analysis and automation is created, with both solutions working together to improve efficiency and support RPA.
Intelligent Document Processing
Intelligent Document Processing (IDP) uses AI technologies like NLP and Optical Character Recognition (OCR) to convert unstructured and semi-structured information into usable data.
IDP’s main advantage for RPA devs is in extracting structured data from unstructured formats like PDF documents, images and email attachments. This enables them to create RPA software bots to automate previously opaque document-centric processes.
Conversational Data Intelligence and IDP are complementary tools. They deliver new automation opportunities by creating entirely new sources of structured data - one from communications, the other from documents such as digital invoices and email attachments. RPA devs can achieve complete, end-to-end process automation, and will never lack the data they need to measure performance or prove automation success.