Healthcare Intelligent Automation Best Practices

Healthcare intelligent automation best practices

In our time developing and building AI-based automations for our healthcare customers, we’ve identified a series of best practices. These practices have helped not only our automation engineers become more effective in their role, but practices that allow our customers to experience quicker value from the automations that we build for them. These practices allow us to build the ideal automation.

Whether you’re an automation engineer using an RPA solution, or a healthcare administrator considering partnership with an Artificial Intelligence as a Service provider like Olive, we think it’s important to highlight the business impacts that these practices can have. In this post, we’ll discuss 3 simple rules we apply to our automation process and the big impacts they have for the healthcare organizations that we serve.

Names matter to ensure the success of your automation

Automations built on an intelligent process automation platform like the one we use to build for our customers consist of a series of logic-based rules and actions. These rules are configured through a web-based interface that then allows you to schedule the resulting automations to run on a defined cadence. While it may seem obvious the number one most important best practice when building automations is none other than – labeling each rule and action effectively.

Here’s why: Although these automations may initially be built by a single engineer, depending on the complexity, they will likely be supported over time by several others like fellow engineers or managers. If you work with an AI-as-a-Service vendor like Olive, you are supported by a Customer Success team, who monitors your automations regularly, optimizing and addressing failures when they occur. Automations are also generally not built overnight. In our case, the engineer becomes very familiar with their customer’s processes as they build them, allowing for iterations to occur as they become more familiar with each business process. Providing descriptive and easily understood labels on each of your rules and actions becomes paramount to ensure the success of the engineer, the supporting team, and the automation.

Try vs. when: Ensuring limited failures & quick error handling

Error handling is important when building workflows to ensure failures are limited and more importantly – that when failures occur, you know why and can address quickly. This is key to ensuring that automations run smoothly and that human intervention is kept to an absolute minimum. After all, the promise of AI and automation is that it will do the repetitive, mundane tasks so that humans don’t have to. This is the promise we seek to deliver everyday to our customers.

Here at Olive, we have a trick for ensuring limited failures and quick error handling – Try and When statements. Here’s how it goes –  whenever you craft an action step in your process automation solution, you wrap it in a Try Wrapping your actions in a Try action ensures that your workflow does not fail, when it doesn’t need to. Here’s an example: When looping through patients, your workflow should ‘Try’ to do something for each patient. Now, when using the Try action, if there is a failure, you should set an error for the patient. Then have it reset to a known point, and have it continue on to the next patient. Now the workflow can continue on, and process the remaining patient list. This can cut the amount of time that your staff spends on specific tasks by 90+%, as they then only need to review and action the errors. The automation has taken care of the rest.

Many workflows will have common failures or optional scenarios that can be handled by a Try, however there are times when a “When” action is more suitable. Let’s say your workflow includes the automation of an action including a browser pop up window, but it does not always include the pop-up. This optional scenario can be addressed by wrapping the action in the proper logic – When. When using a When action, your automation will check the page for the pop up, but will not error out if no pop-up exists. It will simply move on to the next step.

Both of these tricks create a more efficient workflow that requires markedly less human intervention and support.

Intelligent automation: Get on the right page

The first step of every automation should be a step to confirm that you’re on the right page within the given system. Before you start building out actions, before you do anything else, make sure you’re logged in and on the right page or file.

Why is this such an important step? By confirming along the way that you’re in the right place and on the right page, you greatly mitigate the risk of unintentionally doing the wrong thing in the wrong place. This is incredibly important, particularly in healthcare where Olive specializes, as we deal with Protected Health Information (PHI). If we entered PHI incorrectly, there could be grave legal and monetary repercussions.

While it adds a little extra weight to your automation, this step is worth it.

In conclusion

Applying these three simple rules to automation builds can ensure efficiency and effectiveness in your workflow. While these are by no means all-inclusive of the practices we employ at Olive, they are foundational.

As experts in artificial intelligence and automation, our mission is to automate the most burdensome repetitive tasks in healthcare operations to reduce the cost of healthcare and ultimately improve the quantity and quality of human life. If you’re interested in learning more, connect with me on LinkedIn or contact us directly at

Investing for Enterprise Impact Whitepaper

A Healthcare Executive's Guide to AI Workforce Models

Read this white paper to understand the total cost of ownership of various purchasing models for AI and automation, so you can chart the right course as you invest.

Get the White Paper

Related Articles

Learn more