For Australia to become one of the best places in the world to build a tech startup, we need to understand the shape, structure, size, and trajectory of the existing sector. That understanding must rest on high quality data.

Data on startups is an area where Australia is sorely lacking. Over the course of 2019, this problem has worsened with at least one major data provider (Startup Muster) ceasing to operate. Unless this paucity of data is rectified we cannot reasonably hope to make good decisions about how to bolster the sector and won’t fully understand its impact or the opportunities a thriving domestic technology sector presents.

FRAGMENTATION

While there are currently numerous efforts underway to collect and aggregate data on Australia’s startup landscape, a lack of integration with existing programs (both public and private) limits the comprehensiveness of the outcomes. Some prominent community efforts remain, despite Startup Muster’s departure. Initiatives of particular note include Western Australia-based Techboard, and Queensland-based Startup Status. Techboard has built a high quality national dataset from multiple touchpoints focused particularly on startup funding and provides bespoke data services for its customers. Startup Status is a non-profit effort looking to map and measure the impact of startup support mechanisms and has built a very broad searchable database of companies and other ecosystem players.

Government efforts to map, measure, and assess particular ecosystems are also common. In Victoria, innovation and startup support agency LaunchVic has undertaken a comprehensive mapping effort over multiple years, with good results. Other state-based innovation agencies have done similar work on an ad-hoc basis, but more consistency, coordination, and better standardisation of methodology would help improve the utility of these efforts. 

The Department of Industry is undertaking its own startup data project as of Q3 2019. A working group has been engaged to map what organisations fall under the umbrella of the term ‘startup’ and what sorts of data points are valuable for stakeholders across departments and industry. StartupAUS has been involved in that initial consultation and will continue to be involved across Q1 2020 as the project moves into determining the funding and methodology required to establish a comprehensive data- gathering effort.

International startup data sources including Pitchbook, CB Insights, Crunchbase, Startup Genome, the Global Entrepreneurship Development Index, and the Global Entrepreneurship Monitor also form part of the picture, as do efforts from global advisory and analytics firms, such as KPMG’s Venture Pulse. Many of the startup-specific organisations do not have dedicated teams in Australia, which often forms a relatively small market for these organisations. Comprehensiveness of their Australian datasets is often limited as a result.

Fragmentation of data sources remains a key concern, with only limited support currently being provided for any aggregation initiatives. Data collected by government agencies as part of grant programs and other support mechanisms remains largely inaccessible, further limiting the connectivity of data sources. And without a clear, standardised definition of a startup for data gathering purposes, divergent classifications often make comparing datasets impossible.

For further reading on this topic, see Defining Startups at p.29.

MAKING BETTER USE OF PUBLIC DATA

At the same time, a range of existing data sources could be put to better use. Large volumes of data are collected by government agencies in the administration of startup-focused programs at both a state and federal level, but very little of this data is made public. Where it is made public, it is often incomplete.

Some relatively comprehensive data sources, such as the Australian Securities and Investment Commission (ASIC) registry, could provide highly valuable data to organisations seeking to measure the ecosystem.

These are often difficult or (in ASIC’s case) costly to access at the scale required for meaningful collective data analysis. To begin to address this, ASIC data that is currently publicly available for a fee should be made freely available online. Access to the broader ASIC database should also be opened up. Removing these barriers would help data-gatherers access and use public information to build a more comprehensive picture. Open access to company data would also improve transparency and efficiency for investors, incubators, accelerators, and public programs managers who support startups.