Last week, the Capital City Foundation (CCF) published “Big Data in the Big Apple: the lessons London can learn from New York’s data-driven approach to smart cities”.

First of all, I am going to say that it is a great piece of work, going deeper than before into how the NYC model could be applied in London.  It is helpful to a cause you would expect me to back.  And I do in large part back it.  My headline point is that different contexts and a 5 year gap mean that we in London should be more ambitious.  We should go further, and as you will see by the end, look not just across the Atlantic, but also out across the North Sea in search of inspiration.

I will be frank at the outset. Much of London’s data exploitation activity has not been as co-ordinated and therefore as visible as New York’s own. There are good reasons for this which relate to the relative complexities of city government and service delivery on this side of the pond, as well as the political emphasis and centrality of data in very different administrations.  That said, the quality and indeed impact of our own city government analysis and data-driven products, in areas like crime, transport and school places, is exceedingly high and easily comparable with New York.  Testament to this is the GLA Intelligence Unit’s recent success in the Innovate UK “An Integrated Future for Cities Competition” and the London DataStore’s shortlisting this month for an ODI/Bloomberg Open Data Award against strong international competition. Trumpet blowing over.

No surprise then that I am strongly of the belief that now is the time to firmly press the case for a London Office for Data Analytics (LODA) to make real the data driven city Eddie talks about.  This case has to have at its core the stated intent of proving the value city data can deliver, and importantly, doing so in a way that attracts the attention of political masters.  It is to be about organisation, and moving data, New York style, to centre stage.

A key question is what to focus on as we move from data publishing to a data exploitation model.  Many public servants and politicians are fixated right now on the further tightening of departmental budgets over the early stages of this Parliament (see the IFS on this here), the need to consider more public service reform (structure and service delivery models), both of which are set against the push for greater devolution of services in cities.  But then there is also the opportunity to better manage infrastructure and services through big data flows and analytics, and to consider ways in which consumers (aka voters) get better deals out of city markets (e.g. energy).  Arguably the latter two were not such a near-term possibility when Mike Flowers was setting up the Mayor’s Office of Data Analytics, but we would be wrong to ignore them, given the analytical clout in a city like London.  A LODA should therefore have a broader overall focus.

But here is the rub – the noble aims in the last couple of paragraphs go nowhere unless we cajole, encourage, indeed enforce the better supply of data.  I have spoken before about the range of policy issues that pay no heed whatsoever to the capital’s boundaries.  It now seems plain wrong to not have uniform, interoperable data at the pan-London level and the strategic city-wide view of key parts of the city policy landscape, from environmental services, to adult social care, parking and planning.

Other forms of what I have for a while called ‘city data’ also need to feature high on the list of desired data that we should not necessarily open up straight away, but the existence of which improves the delivery of services and life in London.

Of course, there are complex technical (as in data publishing) issues which make data harmonisation in a city of 33 councils, numerous utilities and myriad other interests challenging.  There are legal issues too, but largely, it is the cultural and behavioural aspects of the city data system that leave us foundering.  This is why a MODA for London not only has to be defined by clearly mandated city-wide leadership, but it also has to take a keen interest in developing the building blocks – standards (including the NII), ontologies, a common city data architecture – of the city data estate.

These need to be encapsulated in a properly agreed data strategy for London, as do approaches which will allow for the constructive and secure sharing of data where for competition issues, data privacy, organisations find a reason to say ‘no’. A LODA should clearly lead on the development of this roadmap.

One major potential difference I see between NYC’s model and a LODA is our ability in 2015 to convene a wider range of experts around a data-led innovation effort.  We are much more comfortable now with the idea that no-one provides the complete solution.  This is borne out and understood in enterprises like the Boston Mayor’s Office of New Urban Mechanics. Google’s recently announced Sidewalk Labs also seeks to engage a wide range of partners to take urban issue solving to scale.  It’s Google, so yes, data does play a part also.  A LODA should therefore be clear in how it is built on a bedrock of data experts, but also that this group will need the pokes, prods, challenges and inspirational pushes from those engaged in frontline service delivery, the money men, more exotic propositions like IoT and robotics experts, and many more besides.  Its leaders should be adept at fitting skills and disciplines like citizen co-design around city problems to ensure innovation success and uptake.

I will finish with that glance out to the east.  Over in Copenhagen, exciting city data approaches are taking shape.  A consortium of partners is setting about developing the first platform on which data collection, integration and sharing is centralised for an entire city.  Now Copenhagen is as different as New York is to London – most noticeably it is smaller than both – but what is impressive is the way in which they firmly intend to link data-led innovation to politically recognised outcomes in areas like climate targets and how they aim to be deterministic (there is that word again) in connecting up talent to answer problems.  Finally, they are seriously examining how data can be monetised under a broad set of market principles.  As I understand it, this is not necessarily about always charging for data; but rather they are attempting to be more specific in quantifying the value of city data.  And why should this be a preserve of some and not all in the data market?  Why would a LODA not consider these issues also?