Animal Advocacy Careers is writing a number of skills profiles to support individuals’ career planning.

The process for developing each profile is roughly as follows:

  • We draw out an outline of the profile, with some key headings of content that we want to focus on.

  • We add comments and relevant details from our previous research

  • We search through existing relevant research, content, and advice from contributors and organisations in the effective altruism and animal advocacy communities, especially from 80,000 Hours, the Effective Altruism ForumAnimal Charity EvaluatorsSentience InstituteFaunalyticsThe Humane League Labs, and Rethink Priorities. We add comments and details from there.

  • We conduct brief Google or Google Scholar searches for a few of our uncertainties. This research, unless otherwise specified, is not systematic and thorough. If we do conduct systematic research on a topic, we will usually write it up separately and add it to our blog.

  • Occasionally, we ask connections for their thoughts, although we’re more likely to lean on information and advice that has already been shared with us. We try to be clear where we are relying on anecdotal information and intuition — though we believe that this kind of evidence can be useful.

  • We conduct 5 to 15 interviews with individuals with relevant expertise. We write up the findings into a shared spreadsheet, analyse the answers, and then add additional comments and quotes to the skills profile.

  • We review the first draft internally, then seek feedback from the interviewees and additional reviewers.



Although we have been connected to the animal advocacy and effective altruism communities for several years, we are conscious that our network and the information that comes to us through exposure to that network likely over-represents certain perspectives. Where possible, we therefore use randomisation to select interviewees.

For all of our skills profiles so far, at least some of our interviewees have been identified through the following process:

  • First, we select individuals that had been identified through our spot-check of 27 animal advocacy nonprofits and who appeared to do relevant work. Sometimes, an existing tab of the spreadsheet (our internal version is not anonymised in the same way as the external spreadsheet) already constitutes our pool of potential interviewees, as with the management and leadership and fundraising skills profiles. In other instances, we have to further narrow our criteria, such as by restricting our pool to particular countries.

  • ​We use a random number generator to select individuals for an interview. If the random number generator suggests that we reach out to individuals who we had already had lengthy conversations with about Animal Advocacy Careers in general or about the subject of the skills profile specifically, then we exclude them and randomly select replacement interviewees.

  • We send invitations to interview to our preliminary list of 5 to 10 potential interviewees, via direct email where possible or via shared connections or LinkedIn where we do not have an email address.

  • If individuals do not reply, we send a reminder where possible (this is usually not possible via LinkedIn).

  • If individuals do not reply or decline our invitation, we invite alternative participants. Across the first four skills profiles, participation averaged 62% of those individuals that we initially reached out to.

In other cases, where this method of randomisation does not seem appropriate, such as if we hope to interview individuals from outside the range of nonprofits that our spot-check focused on, we seek out recommendations from a small number of individuals who we believe have expertise on these topics. We then contact all recommended individuals, as long as they seem relevant to our needs, or randomly select interviewees from the recommendations.

For both the “Growing the animal advocacy community in countries where it is small or new” and “Government, policy, and lobbying” skills profiles, for example, we used the randomisation method for a few interviewees. We also reached out to recommended contacts for a further five or so interviewees. We used this second method in order to reach activists in national (as opposed to international) organisations in the former case and people working in public policy or political roles in the latter case.


Questions are selected based upon the needs of the skills profiles in general and any particular questions of interest to specific profiles. The questions are designed to include both open-ended questions that enable the interviewee to share their views on the topic and more specific questions.

As far as seems possible within a natural conversation, we seek to stick to the pre-prepared wording of the questions, to reduce the chances that answers vary due to changes in the question. The order of the questions is not fixed — questions are never asked and answered exactly in the pre-prepared order.

Where questions have no answers recorded, this is almost always because we did not have time in the interview to discuss that particular question; it is very rare that participants have nothing to say on a question that is asked.
As participants speak, we seek to write down their wording as accurately as possible — with direct quotes where possible, or with summaries and paraphrasing if they speak too quickly to write down everything that they say. Wording that is written within quote marks is not necessarily entirely accurate, word-for-word, although we would guess that at least 80% of the wording is exactly accurate.



Once we have conducted all of the interviews, the findings are written up in a spreadsheet that facilitates comparison between the answers of various interviewees to the same questions.

The spreadsheets from our profiles so far are:

The answers are analysed qualitatively and relevant information is added to the skills profiles. Sometimes, individual interviewees are referred to only in the footnotes, listed as the sources for various suggestions and impressions. Of course, if an individual isn’t listed in the footnote, it doesn’t mean that they wouldn’t agree with something if asked directly — it just means that they didn’t emphasise that particular point unprompted. In other cases, phrases, sentences, or paragraphs are quoted in full.  For the management and leadership interviews and growing the animal advocacy community in countries where it is small or new, some additional quantitative analysis was undertaken, as is visible on the second tab of the spreadsheets.

Thanks to all of our interviewees for their help with these profiles. Thanks also to those who provided additional feedback on our careers advice services, including Brenton Mayer, Michelle Hutchinson, Lara Milwid, Erik Hausen, Karolina Sarek, Brett Thompson, Catherine Low, Mattie Toma, Haven King-Nobles, Tom Billington, Tom Beggs, Jacy Reese Anthis, Juliette Finetti, Michael St. Jules, Sofia Balderson, Annabella MacIntosh, and Sarah Lux. Any views expressed are our own, unless otherwise stated.