Do you live in a SmartCity? Our guess is that if you live in a relatively big city, it is very likely “smart”. There are many definitions for what makes a SmartCity, but according to Cisco and the Center for Cities, they are urban areas that are taking advantage of already existing information and communications technology (ICT), such as smart meters or traffic control cameras. The cities then integrate the data gathered from ICT into a single virtual platform and use it to increase service efficiency reduce costs, and generally enhance quality of life within that area.

Simply put, a city becomes a SmartCity when it employs data collection and technology to enhance the services it provides to its citizens.

In today’s knowledge-based, data-driven world, hyper-connectivity can be your best ally. In the transportation sector, for example, emergency services in some jurisdictions are connected to the traffic lights’ operating system and can override it in order to get to the location of an accident quicker. Machine-learning tools can use CCTV data to count vehicles, determine speed and establish patterns to predict traffic volume and road usage. Over time, the historical data, if gathered consistently, could even predict where and when accidents may occur.

Similarly, gaining insights from identifying patterns using already existing and available data is exactly what the asset management industry is striving to do.

In our DataTech whitepaper, we explain how machine-learning tools can make fund servicing smarter. Funds are already collecting a lot of data for regulatory filing purposes. But what if, for instance, a certain item is manually updated on a regulatory report the same way each month? Wouldn’t it be a lot more efficient if the system could detect this condition and suggest an alternate data source or aggregation method to prevent manual intervention in the future? Or, if a fund changes its primary investment strategy, for example, the system could suggest a list of disclosures to be filed based on filings of similar funds. These actions could be automated in reaction to changing regulatory filing deadlines or changing regulatory requirements by simply looking at historical filings, or even historical user-based activities. As long as the technology has access to the historical data and can detect a pattern, the possibilities for self-improvement are endless.

And if we dare look further, since machines are already collecting, cleaning and transforming raw data into actionable datasets, why can’t we enlarge the scope of their connectivity and give them access to more actionable information?

Take an object as basic as an alarm clock. You can set an alarm clock to ring at the same time every day. You can even set special rules for it to ring at a different time on the weekend or on special days of the week in order to get an early start at the gym, for example. But what if your alarm clock was connected to your calendar and could make exceptions to its initial setting to wake you earlier so you get to the airport on-time? Better yet, what if it was connected to the weather and on a snowy day it wakes you up 20 minutes early because, historically, it knows that traffic was bad on those mornings and you need the extra 20 minutes to get to work on time.

With machine-learning tools, the accumulation of data over time allows for any system to spot trends and patterns, be it your alarm clock, the city’s traffic control center, or your fund processing system. Machine intelligence will, of course, always require human intelligence input to validate the patterns and set the policies and rules, but it can analyze the gathered data, identify the patterns, and provide information that human operators would not have been able to see, or at least not as quickly as with technology. And the more data is collected and acted upon, the smarter the technology will get and in the world of fund processing, that could mean processing more transactions at a near-zero cost.

So, the question is, how smart can your fund be?