The future of AI: Is ‘infusion’ the key to data democratisation?

2022/09/12 Innoverview Read

Sisense defines infusion as the practice of incorporating data and insights into end-user business applications. “Infusion is all about putting decision-supporting insights into a product in a way that feels native. It’s accessible. And it’s far more interesting,”  Scott Castle SVP of Product at Sisense says. 

Typically, a BI tool works by pulling data together to help end users draw their own conclusions. They aggregate data, slice and dice, figure it out, come to the insight and then, take action. 

Whereas, infusion speaks towards broadening perspective on what embedded analytics means to include more than just a chart to figure out. An example is the Apple Health Rings on Apple watches, Scott says.

“One of the cool things about this application is that when you’re walking, and it’s counting your steps; it won’t show you a chart. It bypasses that to provide an insight like, ‘Wow, you’re taking twice as many steps as usual; keep it up,’” he says.

That’s infusion. It allows a user to internalise that as opposed to a chart where the user would have to figure out what the information provided meant.

“Infusion is a bigger, broader, more inclusive term. And fundamentally, it’s vital to reaching the billions of underserved knowledge workers — all 90% of them for whom ‘drag and drop BI’ is not really self-service at all,” Scott says.

Three real-world examples of infusion:

1. Decision support: Applications like Outreach use embedded analytics to help users determine the best time to send an email for it to get read. It works by taking the insight, putting it in the UI, and guiding users to the best answer.

2. Connecting users to data: Building a plug that goes from your BI system into Excel, PowerPoint, Google Slides, etc allows users to access the data they need in a tool they already know how to use.

3. Data literacy: Teaching people to use data in their every day, minute-to-minute decision-making is data literacy. One of the things you can do with BI systems is to take a presence app like Slack or MS Teams and attach an NLQ on top of it. That allows you to write a query and the system comes back with a quick cut of data to see if it’s what you’re looking for. Basically, it allows people to start toying with data literacy.

Getting infusion right

Gerimedica, a multi-disciplinary Electronic Treatment Record SaaS platform company that serves the aged-care market, is getting it right. With around 60% of the market share as well as strong relationships with universities and the government, Gerimedica recently decided it was time to evolve and level up its offerings. Its customers were ready for something more. 

“They wanted to connect not only with our applications but their care, staffing, and financing applications as well as Salesforce and Microsoft Office too,” says Hamza Jap-Tjong, CEO and Co-Founder of Gerimedica Inzicht, a Gerimedica subsidiary.

However, as Gerimedica further discussed this with customers, it noticed that, to them, BI had become synonymous with dashboarding. “But BI isn’t synonymous with dashboarding. BI is about bringing intelligence back to your business,” he points out.

Bringing intelligence back to business

Gerimedica knew it had to rewrite the narrative. When it initiated the rollout with its customers, the first thing Gerimedica asked them was, “Who needs what, when, and how?” Hamza regards the “how” as the most important.

“We serve doctors, nurses, and other healthcare providers like psychologists and physical therapists. They cannot incorporate viewing dashboards into their daily practice while providing care for patients,” Hamza says.

“But they still need data to drive their decision-making. So, instead of making them adjust to BI; we make BI work for them by incorporating insights and business intelligence, into their natural workflow. We infuse data from the Sisense platform within our user base.”

For some, that looks like a notification where an alert about something that requires attention is sent. For others, it looks like an email report. However, for the majority of users, integrating data from the Sisense data platform within Gerimedica’s user base looks like collaboration via Microsoft Teams — the biggest in its user base.

Invisible insights infused

Gerimedica leverages BI on a daily basis to help customers, without them ever needing to log into Sisense. In fact, roughly 70% of its users have never even seen Sisense. They’ve never seen a chart. They just have the insights they need in a format that suits them best.

The company also creates ward overviews and sends roughly 350 of them every day to customers. This was initiated to save healthcare providers precious time. Prior to creating word overviews, nurses on the night shift had to copy and paste information from Gerimedica’s system or other systems into a Word or Excel sheet every day. It was taking them approximately 30 minutes for each ward.

By integrating that need into the Sisense platform, Gerimedica now provides nurses not just a dashboard, but an overview in the format they desire and that makes sense to them. Even more, it saves all organisations combined roughly 175 hours of administrative copy and pastes each day across wards. On a per organisation level, that equates to a time savings of 5-8 hours per day.

Serving the underserved means better outcomes

While some of the healthcare providers are what Hamza calls “power users” — those that are able to create data models and dashboards themselves or use dashboarding to create charts, drill-downs, and explanations, he says the majority (95%) aren’t.

With Sisense, more healthcare providers have access to insights, leading to better outcomes. Now they can see the total number of patients admitted and predictions, detailing the number of patients that will likely be discharged. They can also determine how much staff is needed in a particular ward. 

“Instead of just scheduling 30 hours for a psychologist and 20 hours for a dietitian and either overshooting or falling short, with the forecasting system in Sisense, we can predict a more accurate need for future scheduling,” Hamza says.

Even more, healthcare providers are able to see how their patients are doing. “When they see green, then everything is fine. When it’s orange or red, that indicates there might be an issue. They know to give those patients extra attention,” he points out. 

“When they click on the patient names, they get a pop-up with a detailed overview of the medical measurements that are important, helping them make better decisions. With the press of a button, healthcare providers can deep dive into the medical record system to find out more about what’s happening with the patient.”

At the same time, Hamza says that the customers who embed dashboards into their negotiation strategies increased tariffs by 2% on average. “That is a lot of money when you consider that an average organisation rakes in roughly €10 million in revenue for this kind of care,” he points out.

“Their expert negotiation skills based on data have resulted in roughly €200,000 of ‘free money’ — money that can be put towards real estate, education, workforce improvements, whatever they want.”

(Copyright: AI News The future of AI: Is ‘infusion’ the key to data democratisation? (artificialintelligence-news.com)