Chapter 2 Summary & Commentary

Building a city’s leadership dashboard from “useless > usable > useful > good to use” data

One service provider for building smart cities is known for being the first company to successfully develop a leadership dashboard (City Decision System) for large cities. The dashboard targets city governors who want to achieve refined governance of their cities through a Big Data decision-making system.  To achieve advanced decision-making, the core issue lies in integrating existing and incremental data as the key indicators to guide city management, optimise management mechanisms, and enhance their problem-solving capabilities.

In the process of refining data from “useless > usable > useful > good to use” to provide data decision indicators, it is undoubtedly much more difficult to build a city’s leadership dashboard than to conduct digital transformation for an enterprise. In the course of my research interview with this company, they’ve raised the following three questions:

CEO Question 1: Organisations want to disrupt their current management style by going data-driven, which can be a long process. There are so many aspects in the process, from management to governance, requiring integration and transformation.  How can we better plan for this?

Herbert : In my opinion, the most significant difference between government agencies and businesses when it comes to planning is in objectives setting.

Return on investment (ROI) is the obvious modus operandi in business appraisal for a corporate, at least in the short term. However, an organisation’s objectives not only need to strike a balance among various issues but also to support the sustainability of the business.

From a data analysis perspective, there are three possible scenarios for measuring the long term

benefits and the balance issues.

  • Not enough data is available yet and needs to be re-collected.
  • Data is scattered across different departments and needs to be processed collaboratively.
  • There are discrepancies between data because there are no uniform standards, so many resources are required for data cleansing.

Corporates can start their project planning from the above 3 points. They must assess the data requirement scoping for important projects to avoid data integration and development delays that could affect the project schedule. It is also good to anticipate the current state of data through experience and focus on data aggregation and development in advance.

CEO Question 2: Is it easier to increase revenue by keeping the current model intact or proactively innovating when serving institutional clients? In some ways, organisational change and mindset change are incredibly difficult metamorphoses. Moving from a quantitative to qualitative change is too long. Sudden breakthroughs in the industry are only likely to happen by chance.

Herbert : Generally speaking, the lower the maturity of technological innovation, or the more extensive the impact, the higher the cost of moving away from the old paradigm. The highest hidden cost is the organisational divide caused by departments and employees not understanding or not wanting to change.

When discussed in the context of the previous question,  it is easier to solve the localised problem quickly when focusing on the objective.  On the other hand  the overall innovation takes time to transform. Both objectives are not contradictory.

For companies, the question is not whether they should innovate but whether the timing is right. Therefore, when doing a needs analysis, companies must consider the environmental factors from a top-level design perspective and not just rely on the project itself to do the assessment.

CEO Question 3: How to bear the costs arising from the trial and error of transformation and build long-term core competencies in a service organisation – through technology, successful cases, or a deep understanding of needs? Can these first-mover advantages be accumulated into value?

Herbert : Trial and error are inevitable in innovation, but it is crucial for service providers to turn past success stories into products.

Productisation is the process of turning project experience into a generic value and output for broader use. It is a way to move out of being a technical porter.

However, I would like to emphasise that data development projects can also be standardised and productised into a data intermediate layer (semi-finished product) to ensure stability and flexibility in data value output.

As for the data intermediate layer connects different data sources by nature and, therefore, possesses the network effect with first-mover advantage.