Lots of people keep bucket lists—things they want to do in their lifetime. Visit Antarctica. Run a marathon. Hike the Appalachian Trail. For those in the social sector, the list reads more like: Reduce poverty. Stop climate change. Ensure equal rights.

Most people don't put “fail” on either of those bucket lists. 

Yet there is growing transparency around failure in the social sector. Over the last few years, funder groups such as Grantmakers for Effective Organizations and the Florida Philanthropic Network have hosted “fail fests” to encourage foundations to more explicitly discuss failure. Entities such as Fail ForwardFail FestivalsF*&k Up NightsAdmitting Failure, and the Case Foundation's Be Fearless campaign promote the power of learning from failure. Chris Cardona's excellent series of blog posts and SSIR’s exploration of failed social enterprises show the sector is talking about failure more than ever.

But it’s not clear whether all this productive discussion about failure has changed anything in everyday practice. How can we talk about our failures more regularly, and in ways that truly allow us to learn and improve?

This is a question we've been wrestling with at the William and Flora Hewlett Foundation for years. Over a decade ago, the foundation hosted its first-annual Worst Grant Contest. Then-President Paul Brest wanted our staff to be more open about failure; he wanted us to talk candidly about what wasn't working so that we could learn from our mistakes. Every year, staff overseeing each program was required to submit an entry for “the worst grant, from which they learned the most.” The grant had to have been completed or reported on during the previous year, and a short write-up of each was included in a memo to the board. Much in the fashion of today's fail fests, the stories were well told and humorous. Staff voted on their favorite entry, and whatever team submitted the “winning grant” was awarded dinner.

The idea was to create a new norm. It signaled that it was OK—celebrated even—to talk about failure. No one got fired for making a bad grant, and it set expectations: Every year, every program ought to have some type of failure to share.

But over time, we noticed some less-productive trends in the entries. Too often, the focus was on what the grantee did wrong. Research we commissioned from academic institutions ran late and missed the mark, for instance, or their organization imploded due to leadership challenges or financial difficulties. Were these the foundation’s failures? Grantee failures? Or both? (The answer is yes.)

Since the purpose of this contest was to focus on our own mistakes (not blame the grantee), we renamed the contest in 2010 to the Worst Strategy Contest. We were the ones developing grantmaking strategies, and so the idea was to emphasize how we failed—how assumptions and decisions we made didn't play out. The contest continued on for a few more years, and while the emphasis did shift to our own mistakes, over time, it became a bit of a hollow exercise. We were putting too much energy into being funny or creating elaborate skits. We focused on what went wrong, rather than on what we learned and what we could do differently in the future. In the end, we wondered whether the tradition was really making a difference in helping us improve our work.

In 2014, we stopped the contests altogether and started trying different approaches to learning and improving in an ongoing way. For example, we’ve each paired up with a fellow colleague at one of our staff meetings and told stories about something that didn't go well; our partner helped us analyze what went wrong and what we could do differently next time. We’ve read Amy Edmonson's article “Strategies for Learning from Failure” to identify the different ways something can fail—whether through poor implementation, unforeseen external changes, or deliberate piloting and testing of a new idea. We’ve brought grantees in to talk about the risks they faced in starting up new organizations, and how they’ve tried to mitigate those risks. We’ve asked staff members to write down and seal up their predictions for where they might fail in the upcoming year, and then revisit those predictions a year later to analyze what happened, how well they identified potential risks, and what the implications were looking ahead. And as part of our annual review process, we ask each program team to talk about how its strategies have progressed—whether the strategies met expectations, what went well and what didn’t, and what the team plans to do differently going forward. We no longer have a single activity like the Worst Grant Contest to point to, but these different approaches to embed learning and improvement into our ongoing work have yielded rich discussions about our expectations and definitions of success, timeframes for change, and tolerance for risk.

I believe these experiences provide some insight into why it’s challenging to talk about failure in a constructive way. Some might say it’s hard because people aren’t comfortable talking about their failures. While there’s an element of truth to that, and it’s worth addressing, I posit that it's also hard because failure isn't the right way to think about our work. Failure is too blunt of a word for the type of work we do. (We've tried on other terms, like “strikeouts,” but most terms are still too binary.)

Given the complex nature of the work philanthropy supports, it’s hard to identify definitive measures of success. And there typically aren't set endpoints for our work, except for those we impose on ourselves. In the medical world, you can know for sure if a patient died, and you can know exactly when it happened. What does it look like for a foundation's work to fail (or succeed)? It’s not always easy to tell, and we have to look at a whole host of measures and signs that point toward health or sickness. Much of it is subject to interpretation, and it is almost never a simple number or a yes-no answer.

In 2008, the Hewlett Foundation committed $500 million over the course of five years to combat climate change. It was a huge investment for us and a risky bet—it involved starting new organizations, taking a comprehensive approach to addressing climate change, and pooling money with other funders. In 2012, when our president Larry Kramer and our board had to decide whether or not to renew that commitment, it was clear that in many ways our first try didn’t succeed. Some even argued it was a failure: Advocates suffered big losses in high-profile policies, it didn’t raise the additional funds we hoped it would, the organization we funded faced management challenges, and many of the participating networks and organizations didn’t collaborate effectively. Others argued that its successes were numerous: Multiple countries made progress toward reducing or avoiding emissions, it fostered international coordination, and it created new organizations that endured.

So was it a failure?

In the end, it didn’t matter what we called it. It was more important to understand what went wrong and what went well, and use that information to decide what we should do next. The foundation renewed our commitment in 2012, investing another $500 million over five years into the initiative, but we, our funding partners, and our grantees spent a lot of time taking a hard look at what happened, analyzing it from different perspectives, and using those lessons to pivot and adjust in the second round so that we didn’t repeat the same mistakes.

I'm not opposed to calling something a failure if it is one. And I firmly support those who use the language of failure as a rallying cry to promote openness and learning. But in the end, failures are spectacles. Failure focuses too much on what happened, and not enough on what to do differently. Let's not get tripped up by the word. Let's stop debating whether or not something was a failure. Instead, let's focus on learning and improving. Let’s be candid about what's not going well, reflect on it, learn from it, and start doing things differently so we can chip away at those bucket lists.