Operating Values at Open Philanthropy

Ownership means taking responsibility, holistically and personally, for our work. 

At some companies, the ideal employee executes the instructions their manager gives them, exactly as specified. At Open Phil, by contrast, the ideal employee seeks to understand the reasoning behind an assignment and how it fits into Open Phil’s mission as a whole, then does a better version of the assignment than what the manager had in mind.

Employees at all levels of the organization are very often in a position to notice that a project can be done better or faster in both big and small ways – or, perhaps, that the project is ill-conceived and shouldn’t be done at all.

Ownership includes:

• Asking questions: doing your best to understand the full context of your work, to the point where you actively believe in what you’re working on, rather than simply doing it because you were asked.
• Upward feedback: making sure your manager knows where you are and aren’t bought in on the work you’ve been asked to do. Managers, in turn, are expected to value and reward upward feedback, rather than expecting agreement.
• Pride in your work: aiming for outcomes and work products that are excellent and pride-worthy according to your own judgment, not just outcomes and work products that conform to the specific instructions you were given. Even better than asking yourself, “does this answer seem right?” is asking yourself, “is this answer the one that I would generate from scratch if I were thinking this all through myself?” If you don’t feel you’re on track to do this, you should let your manager know and make changes to the project and/or your role as needed.
• Valuing your time: constantly asking whether there’s a less time-consuming way to get the desired outcome. In our view, working more hours rarely increases productivity by more than 25-50% (and even this is rarely sustainable), whereas making good decisions about what steps are essential to complete and which can be skipped, approximated or shortcutted can often result in several-fold (or better) improvements in productivity.
• Focus on the mission of improving the world as much as possible. The best employees consider the mission more important than Open Phil the organization, and more important than their standing within Open Phil.

Openness means constantly seeking, incorporating, and sharing new information. 

Openness includes:

• Giving and receiving feedback. We strive for a culture where feedback is open, honest, clear, direct, and frequent — while also being delivered respectfully and supportively — across all levels of the organization. 
  • Critical feedback can be challenging to give and receive, but we consider it essential. Particularly important is “upward feedback” — feedback that people give to their managers (and their managers’ managers, etc.) We recognize that by default, it can be uncomfortable to give critical upward feedback, and we strive for an environment in which such feedback is welcomed and rewarded.
  • Positive feedback is essential as well. Positive feedback sometimes feels “optional” in the sense that there’s rarely a clear change in actions or policies that results from it. Unfortunately, this can lead to a dynamic in which even people doing excellent work get more critical than positive feedback, which can lead them to a skewed picture of their performance and value. Most managers at Open Phil are currently falling short at giving enough positive feedback, and we’re working on improving at this.
• Checking in: Open Phil tends toward a collaborative and iterative working style with frequent checking in and course correction (though the specific frequency varies by the role and by how long someone has been at Open Phil). The practice of checking in frequently with relevant people applies to everyone at the organization, including the CEO; the goal is not to maximize control of managers over their reports, but rather to save time and lead to better output by getting outside perspectives early and often.
  • Openness to errors in our worldview or high-level strategies. We strive to be open to the possibility that our basic assumptions about philanthropy, philosophy, or how the world works are mistaken, and open to the possibility that programs or grants we fund will be failures. We value it when people seek to be open and accepting of whatever is true, even when resistance to or suppression of information about failure would make their work or worldview look better.
• Risk and mistake tolerance: we believe that failures and mistakes are inevitable if we’re taking appropriate risks, challenging ourselves, and appropriately “cutting corners” to make decisions efficiently. We try to judge people by their long-term value-added, not by the mistakes they might make along the way. And we try to judge decisions by whether they were the right decisions given the information available at the time — not just by how they turned out.
• Making space for new ideas: while we ultimately want to act on well-vetted views, we also need the ability to explore new ideas even when they might be nascent and/or highly unconventional. We encourage people to distinguish between “idea generation mode” and “idea evaluation mode” and to make use of both. We encourage people to engage with unfamiliar arguments on their merits, even if they appear to have unusual implications.

Calibration means thinking and communicating clearly about our state of uncertainty: recognizing and flagging what we know vs. what we don’t. (We mean something broader by “calibration” than the narrow sense described here.)

Calibration includes:

• Transparent reasoning: giving a full and honest picture of what we believe, how confidently we believe it, what our reasons are, and why one might reasonably disagree. For example, we try to use forecasting practices (such as giving quantified probabilities) to communicate about our state of uncertainty.
• Quantitative and systematic frameworks that can help locate key assumptions and disagreements.
• Empirical and logical reasoning. For important decisions, we seek out evidence and arguments that bear as directly as possible on key decision-relevant questions, and we aim to ground our thinking in documented empirical facts and logical arguments that we can follow directly. We are unlikely to put much weight on a study’s conclusion unless we understand the methodology that was followed, the logical and empirical content of the study, and how strongly it bears on the decision before us. We don’t necessarily defer to the views of credentialed people on a topic, and often seek to understand at least part of their reasoning.
• Weighing others’ views and deferring as needed. At the same time, we won’t always be able to understand the full reasoning behind someone’s view; sometimes we need to weigh their view heavily nonetheless, especially when they’re positioned to see something that others (including their manager) don’t. 
  • An example of this is the endorsed/approved/discretionary framework. The CEO and Managing Directors often approve grants that they don’t understand or agree with the case for, due to recognition that the person recommending the grant has expertise and context that they lack.
  • We try to distinguish between “impression space” and “belief space.” “Impression space” refers to “What I would think if I weren’t weighing others’ views at all,” and “belief space” refers to “What I believe, considering all factors, including others’ views.” We do this in the hopes that it will allow us to put appropriate weight on others’ views, while not losing sight of disagreements and while maintaining our own independent thinking.
• Accurate self-assessment. We strive for an accurate picture of each employee’s strengths, weaknesses, and room for improvement, so that we can put the right people in the right roles. Employees that have, and present, an accurate self-assessment are helpful in this endeavor, and more likely in the long run to end up in roles where they can thrive.

Open Phil strives to be an environment where anyone who embodies the above values can do their best work and succeed. 

Inclusiveness includes:

• Avoiding Bias. We strive to avoid bias (including unconscious bias) in our hiring and workplace environment. We ask that all employees consider which behaviors of theirs might be implicitly rewarding or punishing characteristics that aren’t directly relevant to the work (an example might be an employee’s talkativeness or lack thereof in a meeting providing a larger update about their ideas or abilities than is warranted). We also ask that employees point out places where they notice a potential bias, such as against coworkers from particular backgrounds or demographics, or other characteristics not relevant to work performance.
• Flexibility. We have flexible working hours, generally reimburse work expenses that will help your productivity, and take other measures to ensure that everyone can thrive here. We believe that many conventions of the workplace – for example, expecting regular in-person office attendance – are not necessarily important to our work, and are not necessarily compatible with prioritizing inclusiveness. Don’t hesitate to let us know if there’s something we can do to help you do your best work.
• Respect. All employees should be able to do their work and be evaluated on its merits, free from any kind of harassment, unwanted attention, or inappropriate behavior. We expect employees to treat each other with respect and let us know if they are experiencing any issues.