Hand-on-heart, do you really know if your local public services are doing any good? Perhaps they are doing harm. Primum non nocere, the ethic of 'do no harm', is drilled into medics at an early stage. Could we say the same about the rest of social policy and programmes without robust evaluation and evidence?

There are many examples of good ideas that backfired. In the US, Dare, the anti–drug programme, was common-sensical: police forces tried to scare kids away from crime by showing them the grubby reality of convicted drug criminals. The result, however, was the opposite. Although widely promoted by the White House and others, the programmes turned out to increase drug use in certain areas. It was only the evaluation that flagged up the problems.

A lot of financial and political capital was invested in Dare, so it took some guts and honesty for them to open up to research and adapt their programmes. Could it happen here? Local authorities don't feel confident enough that evidence is at the heart of decision-making. Fewer than half of senior public service managers reported in our YouGov commissioned survey that their authority makes 'robust, evidence-based decisions' all or most of the time.

And the same seems to be true of charities. New Philanthropy Capital found that nearly half of small charities with incomes below £100,000 do not measure their impact at all. They are effectively winging it, with no formal measurement of whether their work is having the desired social effect.

The culture is changing

New structures are springing up to make it easier to get evidence-based thinking into our public services. A network of What Works centres has been launched by the Cabinet Office with partners. Covering over £200bn of public spending, the What Works network will drive evidence into social policy, programmes and practice. As well as a new centre on local economic growth the network will bring together established bodies such as National Institute for Health and Care Excellence and the Educational Endowment Foundation. We have also seen the launch of a 'knowledge navigator' to broker stronger links between local authorities and universities.

These are going to be game-changing institutions that will make it easier to find and apply what works. But making sure we demand and value this evidence in the first place is vital. In the Alliance for Useful Evidence report – Squaring the Circle, launched in May this year, Derrick Johnstone argues that we still need to nurture a culture among local leaders that demands evidence. The report suggests things can be improved through building on what is already there, such as the rich data already collected.

The report also highlights the many good news stories that tell of how evidence has improved local services. Birmingham city council, for instance, has treated evaluation of early intervention programmes very seriously, in seeking to establish what really does work through use of randomised control trials. These are substantial experiments which compare the outcomes for groups of assisted children against others who do not take part in the programmes, being used to tackle issues such as 'troubled families'.

There is still much to do to foster a culture of evidence among decision-makers. Martin Reeves, chief executive of Coventry city council, says in his preface to Squaring the Circle that marrying evidence with the real world of practice is a "contact sport". Yes, it's massively hard to apply research in the messy real world of local politics, media and misinformation. But in years to come we will ask why on earth we ever thought it acceptable to fly blind when delivering services. As cabinet minister Oliver Letwin put it at the launch of the What Works centres: "A decade from now, we will wonder how we ever did without it."

This blog first appeared on Guardian Public Leaders Network.