Cold hard data will not put breakfast on the table. It will not be a source of comfort and advice when there is no one else to talk to. And it will not put a roof over anyone's head. Yet the Government's social investment approach - and in particular the greater use of data - could revolutionise the way social services are funded and delivered.

In a pre-Budget speech, Prime Minister Bill English made it clear his Government plans to do social services differently. In his words, "Too often, past governments have judged success only by what they spent, rather than what difference that spending made to people's lives."

At its core, social investment refocuses the measure of success from how much is spent to what is achieved. Data and evidence are used to predict the people most likely to suffer poor life outcomes in order to target and tailor social services.

Data will also be collected to measure the effectiveness of social services to recognise the programmes that are making a difference.

If you think "social investment" is just a fleeting trend in the world of policy wonks and data analysts, you would be forgiven. But actually, social investment should be, and increasingly is, gaining the interest of everyday New Zealanders and those who want to better serve their communities.

Possibly the most well-known aspect of the social investment approach is the calculation of actuarial liability to the welfare system. By looking at what individuals are likely to cost the public purse the most in the long run, policymakers are able to get an idea of where government spending can make the most difference.

To the cynical eye, this may sound like a rather dehumanising approach to social services. People, after all, are surely too complex to be reduced to a series of data points and fiscal liabilities. But it is important not to lose sight over just how transformative the approach could be.

Big picture data and the complexity of human experience are not polar opposites. In fact, social investment is better able to respond to complex and differing needs.

Currently, there is little accountability for when social services actually improve the lives of individuals. It is assumed, but not known, which services will be most effective for which groups. By measuring the effectiveness of social services, data can help pick up the discrepancies between what a policy analyst in Wellington thinks will work and what is happening on the ground.

That is not to say the data collected can act as a crystal ball. For example, government agencies have recently identified four key indicators that are associated with suffering poor outcomes late in life. These include having a parent with a prison or community sentence, being mostly supported by benefits since birth, having a mother with no formal qualifications, and having a Child Youth and Family finding of abuse or neglect.

Those risk factors are not determinants, 35 per cent of children classed as higher risk are projected to not experience any of the poor outcomes identified. Meanwhile children showing no key indicators, or just one indicator, still make up more than half of all children who are expected to have poor outcomes.

If all social programmes were based on those key indicators, and similar forms of risk assessment, then it is likely many kids would still fall through the cracks.

This might sound like a case for more universal support and services. But such schemes assume Government is better at purchasing care for families than letting families spend their money as they see fit.

It is true that Government support might not be reaching the populations it needs to reach, but at least with the social investment approach the Government will know that. At the moment, it knows it doesn't know (and conversely, doesn't know what it doesn't know).

There are already kids slipping through the cracks.

The programmes to address the needs of "at-risk" populations might fail to improve the lives of those they were designed to help, and there might be unintended consequences for targeting certain outcomes or funding certain programmes over others. But again, with the social investment approach, at least the Government will know when it is failing.

The best part is that the public will know too. The Government is not just saying it will change lives, it is collecting and publishing the data so we can judge for ourselves.

Data will not keep a single house warm. But social investment can apply some heat to Government and government agencies to produce the results they seek.