80,000 Hours wants to see more people trying out recruiting

Tl;dr

I’m excited to see more EAs try out recruiting for EA projects. 80k doesn’t “have it covered”, and we are keen to see more people and organisations test their fit with this kind of work.
Recruiting has low “barriers to entry”: you need to find a hiring manager who wants your help, and you need to have the time available to help them.
Work hard to understand what the hiring manager is looking for.
You don’t need a mountain of centralised data in a single CRM to get started. You can start with referrals, spreadsheets, emails and conversations.

Aims

The aim of this article is to:

Encourage more people to test their fit with recruiting, and reduce the extent to which 80,000 Hours is inadvertently “crowding out” other organisations and people from working in the recruiting space.
Communicate some tentative “lessons learned” from our time recruiting in 2019-2020.

Historical context

In 2019 and 2020 I worked on a team with Peter McIntyre at 80,000 Hours, helping organisations working to solve some of the world’s most pressing problems to identify candidates to hire. Below are some reflections and lessons learned from that period. At the start of 2021 we put on hold much of our efforts to help organisations hire in order to focus on increasing 80,000 Hours’ capacity to have conversations with people who applied to speak with us. (We are hoping to increase our capacity on recruiting again in 2023.)

We did ~534 calls with potential candidates in 2019 and 2020. We generated lists of names for roughly 174 roles. We spent just a few minutes on some of those lists, and up to a month on others. We think something like 23 people from the lists we generated were later hired by organisations, though we make no claims about counterfactual impact in this article.

I’ve written this article for people who are considering trying out recruiting, but haven’t done much recruiting before. I’ve put the article together pretty quickly, and expect to update parts of it in response to questions and comments, so please do fire away in the comments section. I’ve posted this on the forum as part of the draft amnesty days.

Lots of hiring managers want help hiring

There are a lot of hiring managers out there who are struggling to find great candidates to hire.
Sometimes that’s because the person they want doesn’t exist, but sometimes it’s because the hiring manager doesn’t have the time to go find the right person.
If the problem is that the hiring manager doesn’t have the time to find the right person, then you may be able to help if you’re willing to put in the time.

Focus on building great models of what hiring managers want in a candidate

Having buy-in from the hiring manager is crucial

The best way to build up a model of what the hiring manager is looking for is to be able to ask them a lot of questions.
Understand how they would trade off between the various attributes they want in the ideal candidate.
If you aren’t able to regularly email/message/speak with the hiring manager during the selection process, it will be a bunch harder for you to find a candidate that will ultimately get hired.
We tried doing recruiting for a few roles in large organisations where we knew someone on the team but not the hiring manager. We generally had a much lower placement rate in these cases.

Give a great experience to the hiring manager

In order to get access to the hiring manager, it’s usually valuable to be as helpful as you can initially. Write out detailed but concise descriptions of the candidates you’re suggesting, put them into tiers, add LinkedIn links, and generally make the process easy for the hiring manager. Once you have more experience you can start cutting corners.

Hiring managers sometimes want the impossible

“I’d like to hire a green unicorn with six legs and five years experience in ML engineering”. Sometimes it feels like hiring managers are looking for someone that just doesn’t exist.
If you’ve tried recruiting in that industry for a little while, you’ll have a sense of the chances of such a person existing and being willing to take this role. Sometimes you need to work with a hiring manager to figure out what “essential” criteria can be made “desirable” in the job description. “Which is most/least important: that the unicorn is green, that it has six legs, or that it has 5 years of experience in ML engineering?” “What if it had just 3 years experience?” etc.
Sometimes these people will exist, but you don’t have the network to find them and attract them to the role. In these cases I usually just say that “My network isn’t great for the kind of candidate you’re looking for”, and when possible point the hiring manager in the direction of someone else who might be able to help them.

On the ground knowledge is valuable

We started off recruiting for a variety of roles across a range of industries. I think this made it harder for us to really understand what hiring managers were really looking for, because they were all looking for something different.
It’s easier to recruit for a role you’ve worked in before and/or in an industry you’ve worked in before. This is because you have better networks in industries you’ve worked in, as well as a better understanding of what hiring managers are looking for. It’s also because you can add more value to the candidate when you are helping them navigate the space.
Professional recruiters usually focus on a single industry for this reason. Effective altruism is small enough that you will probably end up getting requests to help hire for pretty different roles on a fairly wide range of projects. You’ll need to figure out when to say ‘yes’ and when to say ‘no’ to requests to help.

You can get a long way by just knowing a hundred great people

You can test your fit with recruiting and get a long way just using referrals, spreadsheets, and talking with people. This seems to me to be similar to the approach that various community builders such as Sydney von Arx use when building out their networks.
So I’d guess that you don’t need a big database of 10,000+ people or anything like that to test your fit with recruiting. And if you did want “easy” access to some useful data then you can buy a LinkedIn Recruiter Lite subscription for around $170 per month and look through people in various EA-related groups (feel free to reach out if you have trouble with this).
Tom Rowlands, another recruiter, argued in a comment on this article that you don’t need much of a network at all to get started. “You do need to be able to elicit/understand what the hiring manager wants, translate that into useful search strategies, and update based on feedback. You also need a decent knowledge of ‘indicators someone could be motivated / EA-aligned / altruistically driven’ for most roles in the community.”

Data is valuable but expensive to centralise

Broadly speaking, there are three things you want to find out about a candidate:
- Alignment—are they interested in solving the same problems as the hiring manager?
- Ability—do they have the right skills to do this sort of role?
- Suitability—would they take the role if offered? (Location, salary, seniority, etc.)
Data from various sources such as the 80,000 Hours mailing list, LinkedIn and elsewhere were valuable in helping us identify people to reach out to. It also helped us prioritise who to speak with.
When we got a new data source, such as people who attended an EA Global conference and were happy to share their data with us, we usually worked our way through the data source making note of the most promising people we didn’t already know.
BUT centralising data is expensive. We started a project to centralise data from multiple sources to help prioritise who to reach out to. We hit scaling problems with the project and ultimately abandoned it after several person-months of work. It was more time consuming than we expected to centralise the relevant data—especially cleaning the data and converting it all into the same usable format. If you’re thinking about embarking on a project like this, I’d probably be happy to chat in more detail about our experience. In general, the fewer fields you are trying to centralise, and the less data cleaning you have to do, the easier a time you’re going to have.
We would have ideally thought harder about the CRM that we used early on if we had known that we would want to invest a bunch in data. We used Crelate, a specialist recruiting CRM, which had all the features we wanted initially, but quickly ran into scalability issues (e.g. you can only bulk edit a couple hundred records at a time, and there are limits on how many records you can export at a time.) Salesforce would have been a more scalable solution, but it would have required a bunch more setup at the start when we weren’t sure how much to invest in the project. There are many CRMs on the market, and I feel pretty uncertain as to which ones to use when. I’d be happy to chat more about this if helpful, and at various points did comparisons between a few CRM options.
Outsourcing is often easier than automation. Initially I had an instinct to try to automate things like finding e.g. the industry that a list of 2000 candidates work in based on their LinkedIns. In the end, because we had an outsource team up and running, it was frequently quicker for me to write a couple paragraphs of instructions for the outsource team, and just have them go through all the candidates manually adding industries.
Overall I’d guess that most groups starting out doing recruiting should invest less in centralising data than we did. We spent around ¼ of our total capacity on centralising our data over two years. (See also ‘Do things that don’t scale’ by Paul Graham.)

A lot of people could contribute valuably in this space

80k doesn’t “have it covered”. We’d love to see more people working in this space.
Recruiting teams are frequently fairly small. It’s not a “winner takes all” market.
80,000 Hours is currently exploring whether to increase our work in the recruiting space, but we want to see other people entering this space too. We want other people to see the fact that we’re considering working in this space as a sign that we think it’s promising for EAs to work in this space more generally. It’s NOT a sign that we think fewer new people and groups should enter this space!

You can just keep investing more time and getting more returns

The quickest, easiest way that we added value was to just send hiring managers lists of leads off the top of our heads. But it turned out that we could just keep investing more time into a search until we had spent multiple months really understanding what the hiring manager was looking for and finding them the best candidate available for the role.
Some of the recruiting rounds I’m most proud of are cases where Peter McIntyre spent multiple person-months on the round for e.g. Office of the CEO at a top AI lab, or Head of Operations at a prestigious academic institute. In these cases I felt like we helped the organisations make great hires that are still in post today.
I’d be excited to see a bunch more recruiters working for organisations helping out with hiring rounds, as I think there is a lot of work that could be done in this space.

Build feedback loops wherever you can

Ask hiring managers for feedback on multiple iterations of candidates. That way you can improve your models of what they want more quickly.
For each search that you do, loop back once a candidate has accepted an offer. If you recommended them, great! Was there a way you could have gotten to that recommendation faster? If you didn’t recommend them, how could you have found them and recommended them?
Feedback times are generally multiple months, so it’s worth being explicit about creating the feedback loops that will help you get better at the skill of recruiting.

Caveats

We were recruiting in the longtermist space. I don’t have a good sense of how well these lessons apply over to other cause areas.
This document was written pretty quickly, so I expect to make edits in response to comments below.

Thanks to Tom Rowlands, Peter McIntyre, Mike Levine, Howie Lempel, Devon Fritz and Michelle Hutchinson for comments. All errors are my own.