Viewpoints

Overhead is not the issue

Andrew Means on why we need to stop talking about overhead and start focusing on the metrics that matter.

I need to get this off my chest. I am sick and tired of hearing about overhead. I don’t even mean in the overhead-ratio police kind of way. I mean at all. It is an irrelevant metric. It is a complete and utter distraction from the measurements that really matter.

I’ll go one step further. I don’t care if 100% of my donation goes to beneficiaries. Again, it’s such an irrelevant metric.

What I care about is what you are producing with my donation.

Nonprofits are producers. They make things. What they make is change. When I support an organization it’s because I believe they are making the kind of change in the world that I want to see, at a price I think is reasonable. They are bringing arts education to more kids. They are delivering sustainable sanitation systems to children in Nepal. They are ensuring that some of this world’s natural beauty is protected and preserved for future generations.

We give to organizations because there is something in this world that we believe should be different. If they are delivering that change at a price I think is reasonable and fair, then what do I care what their overhead rate is?

I know some people would say, if they are delivering that change but have a 40% overhead rate, couldn’t they do 33% more good if they reduced their rate to 20%? But that assumes that there is no (or only minimal) connection between the amount they spend on overhead and their ability to create change. And that’s clearly untrue. We’ve artificially separated elements in the equation that were  never meant to be separated.

The reason the overhead issue upsets me so much is because it draws  attention away from what really matters. It can subtly make us think that nonprofits are organizations that perform activities and that the goal is for them to perform those activities as efficiently as possible, as defined by doing the most for the least amount of money.

Yet what matters are not measures of efficiency but measures of effectiveness — how well organizations can turn dollars into change. We’ve tricked ourselves into thinking that we can learn any pertinent information about this from an overhead measure.

A marginally better approach is the cost-per-output measure that I’ve seen some starting to discuss. But this metric also has some very negative incentives. It encourages organizations to find the most convenient and least challenging participants and serve as many of them as they can through an intervention that costs as little as possible.

The ideal approach would be a cost-per-marginal-outcome measure, as this is ultimately what we care about. Nonprofits should exist to deliver outcomes — change in the world that wouldn’t have existed otherwise. This is why the marginal outcome component is important. We don’t just want to track how many people found jobs through our workforce development program, we want to track how many got jobs that likely wouldn’t have otherwise. Tying this outcome to dollars is also important so that we know how to efficiently allocate the resources we have across many organizations to achieve the maximum impact.

So please. Can we just never bring up an organization’s overhead ever again?


Does your organization use overhead as a metric? Have you found a better way to measure impact? Chime in below with a comment. You can also connect with Andrew Means on Twitter or LinkedIn

To stay up to date with the latest from Markets For Good, sign up for our newsletter and follow us on Twitter. Better yet, become a contributor!

 

Comments

3
  1. Dan Weiss says:

    This is a great contribution to the conversation on this topic, Andrew. Sadly, we need to keep talking about it to overcome this destructive myth.

  2. Agora for Good says:

    Absolutely agree. Think it’s important to also call out that different types of interventions can (and should) have very different cost profiles – catalytic risk capital for researching a new solution v.s. stable capital for distributing a proven solution, for example. If a donor cares about getting a good ‘outcome per dollar’, the type of funding you provide should mean you are looking at those measurements very differently.

    1. Andrew Means says:

      You’re absolutely right that different intervention types have different cost profiles. We need more data transparency to begin to better understand the ROI of these different intervention types as well.

Comments are closed.