80k supporter update - bout 4 review and updated team values

94 views
Skip to first unread message

Benjamin Todd

unread,
Aug 19, 2018, 1:17:58 PM8/19/18
to 80k_u...@googlegroups.com

Hi everyone,


Here’s our update for the last 3 weeks and the end of bout 4. (See our plan for bout 4 in the last update.)


Summary metrics


Financial

All figures are approximate.


Cash reserves (end July 2018)

$1,967,572

Costs last month (July 2018)

$87,681

Cash reserves divided by costs last month

22

Months’ runway (rough estimate based on expansion budget, including commitments to donate, end July 2018)

11.8



Staff


# of salaried employees (FTE, excluding freelancers)

8.0

Freelancers

14 (2.57 FTE)

Morale (on a scale of 1-10, full-time staff average)

8.2



Note on runway: In Dec, we project we’ll have 7.5 months’ runway on our expansion budget, compared to a target of 6, so we’re on track to slightly undershoot our expansion trajectory.



Note on morale: The figure is pretty stable because it’s the average across all the staff. Individuals tend to fluctuate between 6/10 and 10/10 but it’s hard to move the mean across all 8 staff. However, that means that, a shift of 0.5, say, would be a significant change. Personally, I focus more on the morale of individual staff, and the direction the mean is moving, rather than overall level. Last week we also clarified the scale to sync up better. We define morale as “how confident do you feel in 80k’s long-term prospects and your place in the team this week?” We defined 10 = excellent / little room to improve, 9 = very good, 8 = good, 7 = reasonable, 6 = reasonable with some issues, 5 = significant issues and so on.



Primary growth metric

Change in total impact-adjusted significant plan changes


Monthly growth rate:

4 weeks on previous 4 weeks (ending 2018W32)

-41%


Note that the total change in IASPC is basically zero in both 4 week periods, which produces the large percentage change.


Annual growth rates:

Period

Change in total IASPC

Growth in TTM compared to this period

Monthly compound rate

Last 12 months

1972.8

NA


2017

2262.2

-13%

-2%

2016

1180.7

67%

3%

2015

664.8

197%

4%

2014

244.7

706%

5%



Monthly growth chart - increase in IASPC per month

Note we stopped recording rated-1 and 0.1 plan changes each month in Feb, which is why the steady blue baseline disappears then. We’ll add an estimate of these in Nov.


Annual IASPC targets




Progress in 2 sentences


Michelle Hutchinson accepted an offer to join the in-person team - she’s the former Executive Director of Giving What We Can and Operations Director at the Global Priorities Institute. We also made our bout targets, but IASPC growth is still disappointing.


How you can help

We’re looking for researchers: Do you know who might be able to write problem profiles and career reviews? We’re looking someone smart who has been involved in the community for a few years, and who has some track record of writing. Please contact Rob if you might (rw at 80000hours.org).



-----------------------Full update -----------------------


Bout 4 review


(See our plan for bout 4 in the last update.)


Bout 4 was short at 5 weeks, but we got a lot done:


  • Priority 1: Hiring

    • We made an offer to Michelle Hutchinson to join the in-person team, which she accepted. Michelle was first hired to CEA in 2012 as the Executive Director of Giving What We Can. She then became the Head of Research Operations at the Global Priorities Institute at Oxford, which she helped to set up. She also has a PhD in Philosophy from Oxford. We’re thrilled to have her on the team, and we think she might be an ideal fit for advising.

    • We have several other exciting candidates in our hiring pipeline who might start working at 80k in the next 6-12 months, including one who we’ve made a trial offer to.

    • We didn’t expect to have already made a hire, so we’re ahead of schedule on hiring overall (though we’re a little behind schedule on hiring to the research team, because we decided to spend more time wrapping up in-progress content).

    • We also significantly developed our process for hiring e.g. we spoke to Open Phil, learned new considerations about salary policy, created a new several hour work test, wrote a new set of interview questions, and created a new doc on why to work at 80k.

  • Priority 2: Improving the site for potential-100s.

    • We updated the problem profiles page, which is a key landing page that gets about 15K new visitors each month. We also got clearer on the biggest weaknesses of the site as it stands, and began prototyping changes to the front page, site structure, and other key pages. We used this as an opportunity for Peter Hartree and I to sync up on site design, so that more decisions can be handed over to him.

    • We completed an iteration of the job board, updating the curation process so that the board will now contain a wider range of positions. We also built the most user requested feature – ability to get email alerts when we add new vacancies – and expect to release this early in bout 5. The job board allows us to recommend specific vacancies within each priority path, and there are about 2000 clicks of job adverts on the page each month, so we expect this to lead to some good plan changes each year. Peter Hartree also handed over much of the ongoing management of the board to Maria Gutierrez (our graphic designer since 2015).

    • Advanced series - I spent almost 30h writing, and have finished the most important article on our top career recommendations. I’ve also almost finished the article on coordination, which contains lots of new material, and two supporting articles. We also improved our understanding of discount rates and the explore-exploit tradeoff.

    • We’ve drafted an article on why to work on AI policy in the US government, with basic info on how to get started. We hope this article will help to turn this option into a standard path in the community.

    • We released 4 podcasts, including: Tanya Singh on operations in effective altruism, James Snowden on GiveWell, Prof Yew-Kwang Ng on anticipating EA decades ago, and Spencer Greenberg on how to be Bayesian in the real world. We also recorded 3 more, including one with the famous Vietnam War whistleblower and nuclear policy expert, Daniel Ellsberg. The podcast has now reached 10,000 subscribers, after running for about a year. We’re pleased with this growth, but think it could go a lot further (e.g. EconTalk has about 100,000 subscribers).

    • We drafted an article about our talent survey and have almost finished a profile on academic research careers.


Other key progress:


  • Analysed a survey of ~100 people interested in AI policy careers to determine the key bottlenecks in the path.

  • PH has been training Roman on engineering and Roman made his first commits to the site.

  • We continue to train the team leads in management and on leading strategy within their area.

  • We coached ~14 people.

  • We’re making an offer on an office in SF.

  • We’ve started applying to green cards with a new lawyer.

  • Attended Nexus Youth Global.



Problems we’re facing - IASPC growth:


The main problem on our minds is that we haven’t had any growth in IASPCs for the last 3 months. This is due to the reasons noted last time, and because none of our recent projects cause plan changes that we’ll find out about before the annual impact survey.


We’re not happy about this and plan to focus much more on getting plan changes during bout 5 (mid-Sept to mid-Nov).


We’ve had under 0.5 FTE working on advising so far this year, but over the rest of the year (i) Brenton will switch back from hiring, (ii) Niel will switch back from content, (iii) we’ll have Michelle and (iv) Peter McIntyre will start working on headhunting. This means we’ll have almost 8-fold as much in-person delivery capacity. Most of the benefits of this will come in 2019, but we hope to see some impact before the end of 2018.


Note that we have over 1,000 people on our waitlist for advising with little promotion, and expect we could easily generate a lot more demand.


We’ll also keep releasing podcasts and a series of key articles, which we think can attract great people for the in-person programmes, and potentially directly lead to plan changes.


Note however that we still expect to be fairly flat over Aug and Sept until we get to this work.




Weekly targets


We made 4 out of 5 of our lead metrics for the bout:




Actual

(cumulative)

Target

(cumulative)

Difference

In person team hiring

(target 440 'hiring points'

by end of bout)

1157

140

1095

Content finishing

2

2

0

Advanced series: % completion

(Starting at 41%, target 54% by end of bout)

54%

54%

0%

/problem-profiles/ page redesign: % completion


90%

90% (beta release)

0

Site redesign for tier 1s: % completion of bout 4 subprojects


50%

100%

-50%


Peter Hartree missed the site redesign for tier 1s completion target due to spending more time than planned on the /problem-profiles/ page and the job board.



Updated team values


We rewrote our team values a few months ago. Here is the new summary (roughly in order of priority):


  • Appeal to long-term impact - does this help us reach our greatest potential impact?

  • A humble, scientific mindset - why might this be wrong?

  • Openness and honesty - default to transparency & take the red face test.

  • Focus - say no to everything that's no the top priority, then obssess.

  • Exceptionally well-researched advice - go one step too far

  • Fun and friendliness - say what's true, kind, necessary (and funny)

  • Self-care and growth - don't forget about the long-term.



Read more.



The plan


In bout 5 starting mid-Sept, we want to focus on growth and getting plan changes. Until then, we’ll work on setting ourselves up for that i.e. continue with hiring, on-board Michelle, clear out misc small tasks, take some holiday, and make plans for bout 5, so that we’re ready to push from Sept until the the Dec holidays.



Have a great week,


Ben


Reply all
Reply to author
Forward
0 new messages