Thinking about Analytics?

Add bookmark

Clarence Ti

Everyone is talking about analytics, and has been for a while. At the heart of it, as a user, we are all trying to gain insights from data or set up some form of surveillance mechanism. This article talks about how my organisation thinks about analytics.

But first some background: Vital was formally launched in July 2006 as a captive shared services centre for the Singapore Public Service, serving initially just shy of 20 agencies in areas such as payroll administration, HR services, finance services (primarily dealing with vendor payments) and learning & development services. Today, we serve over 100 distinct government agencies in the original service lines, as well as new areas such as asset management and travel management. Today, over a million transactions pass through our hands, organised along close to 600 distinct lines of service, and our team of close to 500 people serve a population of 80,000 public servants.

Performance Analytics

This family of analytics is familiar to many shared services operators. Figure out what you want to measure and then track it. For a shared services centre, this is usually volume (to determine workload and staffing levels), accuracy (to track errors), and timeliness (to track lapses). The machinery to track these can be amazingly complicated and even complex. We have over 600 lines of services. Imagine if every line carried just one datum for volume, accuracy and timeliness. Then multiply that by 600, and then multiply it by the number of customers each line serves. You’ll soon be in the arena of tens of thousands of indicators … to be collected, reported, and analysed each month. Usually, we try to find ways to visualise these data so that supervisors and account managers can zoom in to the hotspots.

Cost Analytics

Another family of analytics is what we call cost analytics. As a government shared services centre, one of our responsibilities is to the taxpayer, so there is an impetus to lower cost of operations. To do that, we undertake projects that seek to simplify processes, to reduce processing time. Sometimes we undertake projects to reduce turnaround time. That is quite different and reducing turnaround time does not always materially affect processing time if we are cutting wait time. Those projects could end up improving customer satisfaction but not reducing cost. What we really needed to focus on were projects that reduce the time taken to process a piece of work, to reduce the time norms. We think a Lean Six Sigma type of methodology would be ideal.

Consumption Analytics

Consumption analytics is an interesting area for us. This is an area of work that looks at reducing the amount of work coming in from the customers. This is counter-intuitive if, as a BPO or a shared services centre, you are looking to increase the revenue of your centre. For us, as a captive shared services centre, we want to reduce consumption. For this we rank the most resource-intensive processes we have. This is simple to do if you are able to multiply the time norms with the workload volume with the charge out rates. Once ranked, we can look through consumption patterns to look for opportunities across customers. Different customers may share similar processes, so we can determine some "industry norms" and try to facilitate cross learning or rethink certain processes so that, as an entity, we consume less. Sometimes these require policy changes. Where such policies can be adjusted, consumption can fall dramatically. Sometimes, it is about taking a risk management approach and not about spending money chasing down low risk events.

Satisfaction and Engagement Analytics

Surveys, surveys and more surveys. Whether it is for customer satisfaction or employee engagement or a mystery customer audit of specific processes, this family of analytics seeks to understand perception. For the more advanced users, the value lies in measuring change management success or to scanning the Internet for sentiment towards the organisation.

Fraud Analytics

An article about analytics is probably incomplete if we don’t talk about fraud analytics. This category is about detection and surveillance. We think about it as applying a family of health screens to check for exceptions or anomalous patterns. The tricky part about this type of analytics is that any portfolio of screens is only as good as its designer, or its analyst. It effectively pits the fraudster against the designer and the analyst. Some have used specific algorithms such as Benford’s Law to screen, while some have relied on audit teams with their own unique audit and compliance tools to screen for fraud.


For any organisation, embarking on the use of analytics begins with a thoughtful process of what data to collect, what business outcomes or risk we want to track, and learning from how we have been blindsided in the past. Some do this through an enterprise risk management framework while others are less deliberate. We will not always be successful in discovering patterns early enough to prevent damage. Most analytics are not predictive in nature. It is hard to predict fraud when the fraud has not taken place but it is possible to risk profile.

At the end of the day, thoughtfulness counts.

Read more about the evolution of Vital: Singapore Public Service SSO (4-part series)

About Clarence Ti

Clarence was appointed as the Chief Executive of Vital on 2 January 2010. Within public service, he has served on the Economic Development Board in Singapore and in the United States, where he was Director of the San Francisco Centre; the Singapore Land Authority, where he was Director of Land Business & Management; and the Ministry of Law, where he was Director of Strategic Planning. During his career, he has served in areas of investment promotion, business & operations management, planning & co-ordination, and legislative & policy reviews.

He holds both a Bachelor of Science degree with High Honours and a Masters degree in Electrical Engineering from the University of Illinois at Urbana-Champaign , and an MBA with distinction from INSEAD. He is also a Certified Financial Analyst.

About Vital

Vital, a department under Ministry of Finance, was formally launched in July 2006, as part of the Singapore Public Sector’s effort to aggregate common administrative services and reap economies of scale to bring about greater business value for the whole of Government. It is the captive shared services centre which currently serves more than 100 Ministries, Departments, Organs of State and Statutory Boards in Singapore. The suite of services includes finance services, human resource services, payroll and claims services, learning and development services and travel management services. Annually, over a million transactions pass through their hands, with nearly 600 lines of services. Vital has a staff of close to 480 people serving a population of some 80,000 public servants. Vital won the Excellence Award in People & Communications and the Honourable Mention Award for Customer Service at the 14th Annual Asia Pacific Shared Services & Outsourcing Awards 2011. Vital first won the Honourable Mention Award for Best New Shared Service Organisation (under 3 years in operation) at the 2007 Annual Asia Pacific Shared Services & Outsourcing Awards. Vital also received SCORE’s special recognition award 2011. Vital won first prize at the 2nd Pacific-Asia Travel Association-WDA productivity awards 2011.