The Cure for RPA Hiccups in the Life Sciences

March 29, 2019

5 minute read

Robotic hand holding a test tube

When it comes to curing RPA hiccups, our diagnosis is two-fold. The good news is that companies have been able to use RPA in life sciences to increase the speed of drug development by automating manual processes such as data collection, data entry and tracking clinical results. The bad news? RPA tools can hit a roadblock when they encounter unstructured data—leaving life sciences companies searching for relief.

Unfortunately, bed rest won’t cure these ailments; organizations need to take action to ensure their RPA tools are running without hiccups, fed by a steady stream of clean, process-ready content. Read on to learn the four big causes of RPA distress in life sciences, and our prescribed cure for each.

Graphic that outlines the four barriers RPA systems face with unstructured data, including data silos, ROT, volume and PII

Hiccup #1: Data silos

It can take years to develop a drug, and during this time, new strategies and the systems needed to implement them will be created. As new data stores are developed, too often they’re segregated and rife with unstructured data – the kind that poses challenges to RPA in life sciences. The problem is significant enough that analysts are predicting almost half of enterprises will soon have RPA buyer’s remorse due to misaligned, siloed usage.

In life sciences, data silos are a particular issue because a key way companies fuel their expansion and keep their new-product pipelines full is through M&A activity. One of the drawbacks is that the acquiring company often inherits legacy data silos filled with unstructured data that can’t be leveraged by RPA tools.

The cure:

Data silos in the life sciences industry can’t be entirely eliminated. Although data consolidation can reduce the scope of the problem, silos are still maintained because of regulatory requirements for keeping patient health data private, and the need for researchers to share data in segregated databases.

The solution to avoiding silo-related hiccups in RPA is to implement automated data extraction processes that clean and convert unstructured data, no matter where it resides, making it usable for automation.

That way, even if data remains segregated in silos, it can still be treated as one usable resource by RPA tools.

Hiccup #2: Redundant, Obsolete, Trivial (ROT) Data

Our life sciences clients tell us a single piece of content can be duplicated 18 or more times as it gets used in normal business processes. The same data may be used for marketing, clinical trials or research—and gets copied into ECMs, saved locally, attached to emails or copied to new documents. Companies that don’t routinely cull their data repositories end up accumulating vast amounts of data that is just plain redundant, obsolete or trivial—no longer vital to the ongoing work of the business. All of that ROT means that RPA in life sciences has to contend with processing huge volumes of unnecessary data, increasing costs and wasting time.

The cure:

Removing and remediating ROT, and surfacing high-value content can greatly increase the efficiency of RPA in life sciences.

The cure for ROT-related hiccups is to use discovery tools, analytics and content classification to separate the wheat from the chaff, before it is ingested by process automation solutions.

Hiccup #3: Simply too much data

Even after dealing with ROT, life sciences companies are still challenged to process the immense volumes of data being ingested all the time in the form of research outputs, imaging reports, doctors’ notes, lab results, electronic medical records and the output from personal medical devices. The data comes in different formats, a variety of versions of each format (e.g. different versions of MS Word), and up to 80% of it is unstructured data. The amount of content can cause RPA tools to slow down, while the wide variety of formats causes hiccups and breaks in the flow that must be manually addressed, defeating the purpose of RPA in the first place.

The cure:

To cope with large volumes of data, enterprises can make use of solutions designed to automatically and continually enrich data, and extract actionable business intelligence.

These solutions reduce the need for manual intervention and they work at scale, helping life sciences companies cope with the inflow of potentially problematic, unstructured data.

Hiccup #4: Personally Identifiable Information (PII) and other sensitive data

Life sciences companies are especially vulnerable to risk when process automation is used to deal with unstructured data that may contain PII—such as legacy content from clinical trials or data ingested from Clinical Research Organizations (CROs). Enterprises face more than just a bad case of hiccups if PII is automatically dealt with by RPA in an inappropriate manner—exposing participants’ medical data to risk, causing compliance problems and worse.

The cure:

PII is often essential information used for necessary business purposes, but it needs to be carefully controlled in automated processes, like the ingestion of a claim or a patient record.

To avoid hiccups, RPA needs to feed PII to the systems and processes that have authority to access it, and also create protected, redacted or remediated versions of the data for systems that shouldn't have access to the PII, but still need access to the rest of the content.

Wrap up

For life sciences companies to utilize RPA tools to reduce costs, increase accuracy, achieve compliance and their increase speed to market, they must cure the hiccups caused by unstructured data. Enterprises need to be able to access their data and extract the important values in a format that won’t break RPA workflows. This means implementing automated data enrichment solutions that will enrich all of their data before it is ingested by RPA tools or any other process automation tool—and do it at scale.

It’s tough to know where to start when it comes to curing RPA hiccups. That’s why Adlib has a range of options designed to help you get a better understanding of your content, opportunities for automation and clean up. Check out our RPA page or contact us today to learn about our Health Check that will help you get a head start on curing your own RPA hiccups – for good.

Don’t forget to share this post