One of my favorite quotes on performance measurement, from Brian Trelstad at the Acumen Fund, is the following: “Culture matters far more than systems. If your organization doesn’t care about metrics, don’t bother to start building systems to measure performance.”

Yet despite what is now fairly widespread agreement around the necessity of creating a culture of learning and accountability if you want to use data to improve, little has been researched or written about how to go about doing this. I hope this post begins a discussion on what organizations have learned works, and perhaps more importantly, what doesn’t work.

My Bridgespan colleagues Katie Smith Milway and Amy Saxton recently released the results of their research on the learning practices of more than 100 nonprofit organizations. They found three ‘gaps’ that prevent nonprofits from realizing the full benefits of organizational learning. The gaps provide a helpful frame for addressing one of the keys to performance measurement: building a culture of learning.

Setting and communicating goals:

This research found that fewer than two-thirds of organizations define compelling learning goals. The success stories in performance measurement tell us they do the following:

• Identify measurement and learning as a top organizational priority, clearly and publicly, and then follow through with leadership actions that demonstrate the commitment. If the CEO and senior leaders don’t visibly use data to make decisions, front-line staff will wonder why they’re being asked to do the same.

• Communicate around one benefit only: how measurement can improve lives of beneficiaries. Though measurement has many other benefits (improved resource allocation, faster innovation, ability to better attract resources, and so on), the exemplars stay focused on what matters most to staff.

• Avoid jargon. Exemplars stay away from language that can be off-putting to front-line staff; they painstakingly make the language of measurement accessible to all.

Creating incentives:

The Bridgespan study found that only half of organizations surveyed created incentives for staff to capture and share knowledge. The success stories in performance measurement tell us they do the following:

• Integrate measurement into job descriptions, performance reviews, and promotion criteria. If you aren’t formally rewarding good behavior, staff will quickly lose interest.

• Supplement by recognizing and rewarding in visible ways. One organization we recently spoke with issues a ‘monthly measurement award’ to the employee who best used data to improve outcomes for a beneficiary in their caseload.

• Recast ‘good failures’ as learning opportunities. Staff will quickly lose motivation if they perceive colleagues are unfairly ‘punished,' for example, for reporting that a new technique they tried did not appear to improve outcomes for their beneficiaries. While using both carrots and sticks is important, the key in using the latter is distinguishing between ‘good failure’ (e.g., experiments) and ‘bad failure’ (e.g., deviance from a prescribed practice). Learning organizations are expert at using good failures to improve. (For more information, see "Strategies for Learning from Failure" at HBR.org.)

Developing effective processes:

In the study, only 40 percent of organizations believed their processes were effective for encouraging learning. The success stories in performance measurement suggest the following:

• Make the process of learning as effortless as possible. Exemplars search for and improve processes that most frustrate their staff, such as downloading reports from data systems.

• Build in the time and space for learning. Learning processes and mechanisms, such as weekly staff meetings to reflect on data or online portals to share best practices, help to embed learning in the daily rhythms of an organization.

• Start small. When trying to engender a learning culture, it is often more effective to work with one, high-profile program or site (that will see the benefits and be your advocate to the rest) as opposed to what one organization described as ‘slamming measurement down the throats of everyone at once.'

One final note: the organizations we’ve spoken with that have gone through a transformation in their culture (from ‘instinct-driven’ to 'data-driven') tell us it takes them five years or longer to fully reap the benefits. Therefore, anticipate the need for a healthy dose of patience.

Are you striving to create a learning culture? What has worked for you? What hasn’t?