Singtel’s data analytics business, DataSpark, has achieved some impressive results, but scaling is hard. Its path highlights lessons on dealing with the challenges facing all telcos building new businesses, e.g. how to govern and manage relationships with the broader organisation, measuring success, and finding the right skills and partners.
Data analytics as a new business
This case study looks at DataSpark, an autonomous business unit of Singtel (www.dsanalytics.com) and evaluates the benefits of creating a separate organisational structure within a telco to provide technology and support for the development of analytics, AI and automation as a new business. It is created after conversations with Shaowei Ying, Chief Operating Officer of DataSpark. The company’s activities include both the creation of internal capabilities and data monetisation capabilities for external customers.
DataSpark was formed in 2014 at a time when not many telcos were actively exploring new data business opportunities. The unit consisted of a small group of data professionals with skills around, particularly, location data. Singtel’s CEO was a strong supporter of leveraging telco data to establish competitive differentiation and therefore tasked them with looking at various location-related external monetisation opportunities. It was considered natural to create internal use cases for the data to defray the cost of the data preparation. In particular, the same mobility intelligence was of use to radio network planners optimising their network roll out using not just congestion, but now subscribers’ mobility patterns, too.
Enter your details below to request an extract of the report
DataSpark’s progress to date
Telcos’ external monetisation units, such as DataSpark, are not yet large enough to split out the revenues in their reports and accounts. However, in the 2018 and 2019 Management Discussion and Analysis DataSpark’s progress was reported to include:
- Activity to bring mobility data to sectors such as transport and out-of-home media in Singapore and Australia
- Partnership in out-of-home advertising with large players taking a data-as-a-service solution to optimise their assets
- Provision of insights including first party enterprise data in the consumer goods sector to deliver new use cases in advertising and retail store inventory optimisation
- Recent support for governments in predicting spread of Covid-19, including understanding the socio-economic impact of the virus.
Service example: COVID-19 insight for the Australian local government
Table of Contents
- Executive Summary
- Two diverging strategies for a small, independent data unit
- Scaling up the data business as an integrated unit
- DataSpark’s progress to date
- DataSpark’s approach to building a data unit
- What services does it offer?
- Go-to-market: Different approaches for internal and external customers
- Organisational structure: Where should a data unit go?
- How to scale a data business?
- The immediate growth opportunities
- Following in others’ footsteps
- Building new capabilities for external monetisation
- Assessing future strategies for DataSpark
- Scenario 1: Double down on internal data applications
- Scenario 2: Continue building an independent business
Read more about STL Partners’ AI & automation research at stlpartners.com/ai-analytics-research/