30 minutes! A recent report from AppDynamics shows that this is the average time lost in productivity due to slow or poorly performing IT systems per day. For any health care employee, this means that you experience daily delays in accessing (patient) information, resulting in longer visiting time of patients. These disruptions in workflows means more time spent on administrative tasks and less time spent on direct patient care.
Anyone can visualize these moments where you want to access information, press enter on your keyboard and then… nothing happens. You click again, and again, and still nothing happens. In web performance terms, we call these ‘rage clicks’, pressing the button multiple times hoping that the system will respond (which usually doesn’t happen). These daily losses in productivity could lead to a negative outcome for patient care and increased work-related stress for health care employees.
To measure or not to measure?
A second of delay can already lead to a 7% loss in conversion for any e-commerce website. Therefore, most e-commerce companies spend a lot of money and resources on making their digital touchpoints faster, and at the same time spend a lot of money on measuring performance to prevent any performance degradation. In health care, IT budgets are similar if not bigger compared to eCommerce companies, yet the average loss in productivity is bigger. Is running health care applications that much different?
Yes, health care is different in such a way that any digital touchpoint needs to be more secure and reliable than an ecommerce website, meaning it’s more complex to build and manage. But, at the same time, the technology used to create these digital touchpoints are similar and so is the way you can measure the performance of this technology.
If we can measure similarly, why does research show that there’s still a daily loss in productivity due to poor performance?
The limits of system monitoring
The biggest caveat is in the complexity of the components used to create a digital touchpoint. For any healthcare system it takes multiple parties, internal departments, and external vendors to create one digital touchpoint used daily. Combined with high standards in data integrity, the complexity created is exponential.
This complexity impacts the way we can measure performance. Standard practice is to use Application Performance Monitoring, which typically focusses on individual applications and their performance metrics. A more recent trend is to shift to Observability platforms like Dynatrace, AppDynamics or Splunk. With Observability, you still focus on the same application, but it allows you to ingest data from multiple sources, including logs, metrics, and traces to provide a comprehensive view of system performance. In short, Observability allows you to collect more data to create a better view of your individual application.
But as mentioned, any healthcare technology is highly complex, with numerous processes, components and workflows that must be carefully managed to ensure optimal patient care and operational efficiency. Can Observability help us with our goal to create one comprehensive view of reality?
Let’s use the e-commerce website as example, where a regular online transaction probably looks like this:
- As a user, you enter the domain name in your browser and go to the homepage of a brand website.
- From there you browse through promotions or view various products.
- When you decide to purchase, you select the products and add them to your basket.
- You go to the checkout, pay and confirm your purchase.
- And then wait for PostNL to deliver the items at your home to complete the purchase.
Now, for the e-commerce brand, their responsibility lies with the performance of the features on their website. But for the customer, the transaction ends when PostNL delivers the products. This is called the difference between features used and a customer journey.
A different example, more similar to health care, is UWV. If you had to apply for NOW during Covid, a process would go as follows:
- As a company you go to the special NOW website and apply.
- Your credentials get checked and verified by UWV.
- Based on this verification, you will receive a confirmation of your application.
- If approved, you are formally registered and enter the NOW protocol.
- Once entered, you regularly update your figures to trace the NOW payout.
For the SMB companies, all the steps equal one flow and thus a single customer journey. For UWV, each step is an individual department within their organization. And each of these departments has their own technology, external vendors to manage, but also works on their KPIs and service levels. For each of these departments it is their task to manage the performance of their application, not the quality of input and output of the entire flow.
And this in and output is where we reach the limit of an observability setup and need to change our approach. Only if we connect all the dots from the viewpoint of a customer journey, we can understand if we are delivering according to service levels. And if any service level is breached, where this degradation arises, who is impacted and who is responsible?
Make the mind shift, from A(pplication performance monitoring) to B(usiness process monitoring)
To really solve the loss in productivity, we need to introduce a new approach to monitoring. This approach is called Business Process Monitoring, the art of tracking and analyzing key performance indicators (KPIs) from the perspective of a customer journey or business process.
The examples from the eCommerce website and UWV show us that monitoring a business process (= customer journey) is essential to stay in control of delivering an optimal customer experience. For those already using Observability, implementing a Business Process Monitoring approach does not involve a major change in monitoring components. It simply combines the technical features of Observability with a birds-eye view approach of your entire landscape:
- Creating a monitoring strategy that focuses on analyzing the entire business process, measuring inputs, outputs, and resources of all components independent of who owns them.
- Create, track and analyze key performance indicators (KPIs) to identify areas of improvement within processes. In the context of healthcare, this could mean monitoring processes such as patient admissions, discharge, and transfer or and medical record documentation.
- Using Open Telemetry standards as a design principle in your monitoring strategy to include data from all components into your observability platform.
- Data collected from technical components you own, will become your data lake. Make agreements with other teams and third parties about the supply of telemetry data (metrics, traces, logs).
- Extend BPM and telemetry data in your DevOps process to prevent degradation to be introduced into production, so called Quality Gates.
30 minutes a day, or 16 days, in productivity is lost each year
If you adopt the Business Process Monitoring mindset, you can start reducing complexity in your IT organization and gain back productivity:
- Real-time visibility into operational efficiency – independent of who owns the components.
- Improved patient outcomes and employee happiness – Identify trends and patterns that may be impacting patient outcomes.
- Cost savings – Reduce costs and allocate resources more effectively.
Are you ready to shift from A to B?