Effective data-gathering for process efficiency
It’s vital to have a strong data-gathering foundation before looking to smart technology to produce executive dashboards, alerts and autonomously-generated instructions.
Businesses operate in the smart technology era and have access to powerful tools for data collection and process automation that were unimagined just a few years ago.
We have the power to use robotic process automation combined with machine learning to autonomously manage processes, resulting in greater efficiency, lower operating costs and better customer experience.
One of my long-time clients is a large bank. Every morning all its staff − and there are more than 40 000 − need to fill in an Excel form as part of a daily COVID health check, and send it to their line managers by e-mail.
This bank has invested millions of dollars in its technology platforms, utilising the latest and greatest technology platforms from the likes of Microsoft Azure and Amazon AWS, so I was astounded to learn that it requires staff members to complete the same document, on a daily basis, and send it by e-mail.
Imagine the hundreds of thousands of e-mails sent on this basis. I can bet you that most line managers do not have the time or inclination to process all these COVID health check e-mails. It makes me wonder what was its thinking and the purpose behind this questionable decision.
There is no way the bank can smartly analyse all the e-mails, create alerts and look for patterns in the data. If it did it smarter, it would have used a web or mobile channel, making it easier for staff and requiring less time from everyone. The data collected can immediately be analysed for patterns on a daily basis.
What patterns could smartly collected data show, in this case? You may, for instance, find that staff from the same location − like branch staff − are experiencing high temperatures and other COVID-related symptoms.
That is a clear indication that something is wrong and immediate action should be taken. In fact, if you use smart technology, you do not need a human to make that decision. COVID symptoms from staff in the same location could trigger an automated message, instructing the affected staff members to self-isolate, while the matter is investigated.
For basic data-gathering and reporting, there is no need for expensive or additional technology.
We need to lay a strong foundation of effective data-gathering before we look to smart technology to produce executive dashboards, alerts and autonomously-generated instructions:Be clear on the strategy.
- What information do you intend to gather?
- Where will you get this information, and how will it be collected?
- Why is the information needed, and what questions are the information going to answer?
- Who will use the information once collected? How will the information be analysed?
- How will any analyses be used (ie, reporting, budgeting, planning and prediction).
Use the tools you have. For basic data-gathering and reporting, there is no need for expensive or additional technology. The likes of Microsoft and Google provide tech that is either free or already part of your licensing (ie, Microsoft Office 365).
Capture the data at source. The best, fastest and most accurate way to capture data is “at source”. Allow staff and customers to enter data through digital channels on their smartphones or using a chatbot on your website. Problems come in when the source data has to be reworked (in the case of written documents that later need to be entered into a system by another person).
Avoid rework. Connected with the previous point, identify any data capturing that requires rework. For example, capturing the data a second or third time into additional systems. The source data can be used to directly submit information to various systems.
Make it easy and keep it simple. The people who capture the source data and those who need to use the data should be empowered. Only the most important data should be collected. Avoid making it a laborious exercise by requiring data that is not important or that has already been submitted. (An example is asking for a client’s name and address. You should already know the address based on the CRM system info. The name can be linked to the address, or other data already in use, by connecting it to the name once submitted.)
Data privacy is key. The Protection of Personal Information Act (POPIA) is now law. A breach can result in up to 10 years imprisonment and a R10 million fine. A breach is when the data is "lost" or hacked or used for purposes other than for why it was gathered. Key here is personally identifiable data (name, telephone number, ID number, address, etc). In the case of Google and Microsoft, the data is safely stored in the cloud. But give consideration to how the data is used in your company. (Who has access and why? Is there data on printed paper that is not stored safely; ie, in desk drawers?)
Change management is important. Those who are required to capture the data should see how it will benefit them. If they see it as yet an additional administrative requirement, then the uptake will be slow. Ensure you keep the front-line staff and customers involved. Ask them what would they like that would make their lives better from a reporting perspective. The roll-out of the data-gathering portal should be done with the right messaging and change management in mind.
Smart automation, artificial intelligence thought leader and management consultant
Johan Steyn is a smart automation and artificial intelligence thought leader and management consultant. He is chairman of the Special Interest Group on Artificial Intelligence and Robotics with the IITPSA (Institute of Information Technology Professionals of South Africa). He writes in his personal capacity. .
Johan Steyn is a smart automation and artificial intelligence thought leader and management consultant. He is chairman of the Special Interest Group on Artificial Intelligence and Robotics with the IITPSA (Institute of Information Technology Professionals of South Africa). He writes in his personal capacity.