Seven principles for a new approach to learning and accountability

In 2019, GO Lab and Collaborate CIC (with Northumbria University) each published reports which explored the move towards systemic and collaborative working locally: Collaborate and Newcastle Business School’s report Exploring the New World and GO Lab’s report Are we Rallying Together? A key question that emerged from both reports was how dominant approaches of target-setting and impact measurement, which are typically based on linear models of causality and control, could be reimagined to enable effective collaborative approaches to tackling complex social issues.

In April, GO Lab and Collaborate CIC hosted a virtual roundtable bringing together public, civil society and academic partners who represented wide array of views on this question. In the roundtable discussion it was acknowledged that an effective model of accountability needs to bring a focus on learning and adaptation alongside the usual focus on value and effectiveness. Current performance systems are largely orientated towards the latter, with sharing of learning within and between providers and commissioners seldom encouraged — for a variety of reasons. These range from a belief that competition drives innovation, to the prioritisation of ‘compliance’ over learning.

As many have acknowledged, this is problematic, as outsourcing through a model that inhibits providers returning their knowledge into the system results in a commissioner (and wider system) which understands less and less about the issue, the service, and the people they seek to support.

What is needed is an approach that brings together ‘value and effectiveness’ with ‘learning and adaptation’ in a context-informed way, seeing these as mutually supportive (rather than oppositional) ambitions.

The seven principles 

Below we identify seven principles that emerged from the discussion and can help underpin such an approach.

1. Start with a systems perspective: broaden your lens to uncover the multiple factors and actors that can help understand and address the issue. This helps surface what really matters as the foundation for making the case for and developing a helpful (rather than restrictive) approach to accountability. Understanding the perspective of people who access support, and citizens more widely, is an important part of this.

2. Build shared purpose: focusing on purpose can be a useful way to ensure direction without over-reliance on restrictive targets. Approaches to understanding different perspectives and developing shared purpose should be a collective endeavour — to bring together insights from across the system and to build lasting trusting relationships as a foundation for working towards the shared goal

3. Focus on enabling informed judgements: use feedback and learning to build the expertise of all stakeholders, including funders, practitioners, service users and citizens, so that everyone makes better informed decisions and judgements about how to achieve the best outcomes in different circumstances.

4. Develop legitimate measures: gather information that enables these informed judgements by understanding what all key stakeholders need and value. As part of this, consider how to maximise transparency and create equal responsibility between funders/commissioners and providers for learning and accountability.

5. Review goals: as well as enabling flexibility in delivery, be prepared to change goals and the way success is measured to achieve the overall purpose as the system learns.

6. Take a context-led approach: develop an approach to accountability that recognises how complex a context is and therefore what is possible and useful in terms of measurement. Alnoor Ebrahim’s work is helpful here. He suggests that choice of performance management systems should be determined by 1) the extent to which there is uncertainty about the cause and effect 2) the extent to which the organisation or person being held to account by the system has control over the outcomes being measured.

7. Take a nuanced approach to evidence: there is rarely a situation where we know all the answers or when we have no existing insight into what might be effective. Rather than viewing evidence-led practice and bespoke approaches as competing perspectives, share data openly and consider how we can usefully build on what is known while also recognising the limitations of our knowledge, and the need for flexibility and adaptation.

Join us at the Social Outcomes Conference 2020

The GO Lab and Collaborate CIC will host a follow-up to the April roundtable at the Social Outcomes Conference 2020, which will take place virtually this year and is free to attend. This session will also invite international perspectives, and will sit alongside other sessions discussing similar issues. You can sign up here.

Following this, we hope to convene regular practice-focused sessions to explore real-life examples of new approaches to accountability and learning, and explore how these can inform practice more widely. One of the key questions we seek to explore with partners is: If we accept that there isn’t a single version of what constitutes a good system of learning and accountability, what are the factors that should influence our choices?

Thank you 

We would like to thank everyone who participated in the roundtable who represented the following organisations: Dartington Service Design Lab, Northumbria University, Plymouth City Council, Gateshead Council, Social Finance, NLGN, CPI, The Cares Family, Easier Inc, Red Quadrant/Commissioning Academy, Centre for Youth Impact, Black Thrive, Office for Civil Society, DCMS, Nesta, iMPOWER, National Lottery Community Fund, Camden Council, RATIO, Barking and Dagenham Council, Institute for Government, Lankelly Chase, Youth Endowment Fund / Blagrave Trust and Sport England.

This blog post draws on both discussions at the roundtable and, GO Lab and Collaborate’s wider work. Views expressed here are not necessarily representative of all roundtable participants.

You can view the original version of this blog on Collaborate