Lessons learned from collaborating on data-driven initiatives

Cross government collaboration has been repeatedly identified as being central to the success of data-driven initiatives.

Attendees at the recent Think Data for Government event in London heard examples of cross-sector collaboration that ranged from protecting some of society’s most vulnerable to using data to inform future approaches to community-led, data-enabled problem solving on the high street.

London High Streets Data Service

In a session dedicated to the topic (pictured), Craig Campbell from the Greater London Authority (GLA) highlighted the London High Streets Data Service, a partnership of 35 local authorities using pooled resources to access big data on high street spend. He said there had been a gap in evidence on how well local high streets and town centres were recovering from the Covid-19 pandemic.

The local councils were using a patchwork of different data sets to understand the recovery. There were also concerns around the cost of paying suppliers to access to that data, and a general lack of coordination or a joined up agenda for research and providing insights on the recovery, said Campbell.

Working with Mastercard and a range of partners, the GLA created a hub which curates third-party data for organisations that might not have the resources to do it on their own. In addition to aggregated and anonymised spending data, the service pulls in footfall data, vacancy rates and even site-specific information, like opening and closing hours. Campbell’s team turns that information into a dashboard for each high street, with insights for its local authorities.

“What makes this program interesting in terms of cross sector, cross government collaboration is the commercial model that we use,” said Campbell. “We collect subscription income from those 35 member organisations, aggregate that…for private suppliers to [use] those insights, which then our data scientists turn into dashboards, tools, reports as well as data sets, which we get back to our partners.

“I think the commercialization, as opposed to a voluntary data partnership, is a really interesting way to secure that flow of value and the durability of the program, and that it raises the stakes with the client organisations.”

GAMSTOP

In terms of a civil society initiative, Hippo’s Simeon Souttar detailed, a scheme to help individuals restrict their online gambling. GAMSTOP that allows individuals to exclude themselves from all UK-registered online gambling websites. The solution provides a central point of exclusion for consumers whilst providing real-time integrations to 150+ UK-licensed operators to check their customers’ exclusion status each time they register or log in.

“It came around as trying to solve a problem where each individual organisation would offer tools to manage safer gambling, but there wasn’t a central point where you could exclude from all organisations,” explained Souttar.

“Hippo have been a partner to GAMSTOP for the last five years, and we have helped GAMSTOP develop the scheme during that time. It solved some interesting challenges in the industry; we see a very large volume of data that is being shared with the 150 organisations. We act as a central service that provides both real time and batch API calls to those organisations.

“So if you imagine the typical journey of someone who’s interacting with the service, they may register themselves on the scheme, and then they’ll be added to a central exclusion register. We hold that register, and then an operator at the point someone is trying to log into their service is checking our register in real time to understand whether they should allow that to take place or to block it at that point of entry. So as that individual is logging in, we’re having an impact to 150 organisations that are dealing with lots of login data in real time. We see a very large volume of data to be shared across those organisations. We’re effectively sending key information about each user, and we’re patching that to our register. We see over 300 million requests to our API every day, so there’s an extremely large volume that we’re managing across that period, continued Souttar.

“And the nature of the service is also quite interesting, because we see peaks and troughs from user demand, specific sporting events, specific periods of year, where we see very large spikes – very large racing and football events caused these big spikes in demand. So we manage that service across that scale, in the cloud to deliver it.”

OpenActive

Elsewhere, Andrew Newman from the Open Data Institute (ODI) discussed OpenActive, a Sport England initiative to standardise data across the sporting sector. Its focus is on reducing health inequalities, particularly in underrepresented groups.

The initiative involves organisations from across the sector, including leisure centers, sporting national government bodies, gyms to individual providers, extracting data from their individual booking systems in a standardised way and make it available to innovators to build new applications that make it easier for people to find, book or provision physical activity.

“The first challenge is the sports activity sector has a really low level of digital maturity compared to other sectors,” said Newman.

The latest phase of OpenActive is designed to help shift the sport and physical activity sector towards greater data standardisation, which will make physical activity more accessible to all, especially under-served communities.