80k supporter update - bout 4 plan and update on annual targets

Skip to first unread message

Benjamin Todd

Jul 25, 2018, 11:52:28 PM7/25/18
to 80k_u...@googlegroups.com

Hi everyone,

Here’s our update for the last 3 weeks and plan for the next bout.

Summary metrics


All figures are approximate.

Cash reserves (end May 2018)


Costs last month (May 2018)


Cash reserves divided by costs last month


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



# of salaried employees (FTE, excluding freelancers)



8 (2.485 FTE)

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


Primary growth metric

Change in total impact-adjusted significant plan changes

Monthly growth rate:

4 weeks on previous 4 weeks (ending 2018W28)


Annual growth rates:


Change in total IASPC

Growth in TTM compared to this period

Monthly compound rate

Last 12 months



















Monthly growth chart - increase in IASPC per month

Annual IASPC targets

Progress in 2 sentences

We’re behind our targets for the year due to having less coaching capacity than planned and recording some negative plan changes. This bout we’ll focus on hiring and improving the online content for potential-100s.

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

Recording negative rated plan changes

At the start of 2018, rather than tracking the number of new plan changes recorded each period, we switched to tracking the *change* in our all time total IASPC each period. This means that if we initially rate a plan change with value 100, but later downgrade it to 10, then in that period we record -90 IASPC. This is why several of the most recent months have been near zero or negative.

Note that recording negative values is not surprising. Many of the rated-10 plan changes, for instance, are effectively people we think have a 10% chance of becoming rated-100 plan changes and a 90% chance of being rated-1. This means that in the long-term 9 rated-10s will get down-rated for every 1 that gets up-rated.

Last month we took the further step of also recording negative value plan change ratings i.e. we rate plan changes as -1, -10, -100 and so on if we think we made their career lower impact, we caused someone to waste resources, or we contributed to a project that made the situation worse. For instance, if we recommend someone apply to an organisation, costing staff time from that organisation, but the person doesn’t work out, then we’ve incurred a cost for the org for no gain, so it’s a negative impact.

Due to some negative ratings and downgrades, we recorded -291 IASPC in May, which explains the drop in the graph above.

This has put us significantly behind our growth targets. Currently, we’ve recorded 321.5 IASPC so far in 2018, but that would have been 611.5 without this downrating. That would have made us close to the dotted line on the graph, which shows our annual target. All this is unfortunate, but gives a more accurate view of our impact. We also remain well ahead of where we were this time last year.

Update on annual targets

Unfortunately we think it’s fairly unlikely (maybe 30% chance) that we make our annual target of 2200 IASPC. This is due to:

  • Not having Peter McIntyre for 9-10 months, meaning our coaching capacity was lower than expected, and didn’t grow from 2017. In this time, Peter could have likely found several hundred IASPC and maybe hired staff who would have contributed later in the year.

  • When we set the 2200 target, we weren’t planning on counting negative plan changes.

  • Our greater focus on rated-100+ plan changes. This is starting to pay off in that we’ve recorded 4 rated-100 plan changes in 2018 so far compared to only two over all of 2017; however, most efforts to get rated-100 plan changes take over two years to pay off, so many of our efforts are not yet impacting our metrics.

  • Our decision to focus on hiring this bout. We’ve thought about it at length, and we think the in-person and research team should invest significantly in hiring over the coming months. It’ll be better long-term, but doesn’t especially help with our 2018 growth targets. Likewise, we think Rob and Peter H should continue to focus on the podcast and website respectively, though this mainly pays off over several years rather than 6 months.

Currently, we expect to record 800-1600 IASPC over 2018. A major uncertainty is the probability of recording another rated-1000 plan change. If we do this, then we’ll probably make the target, but we think the chances only 20-30%.

We’re pretty disappointed about this, and plan to spend bout 5 heavily focused on getting plan changes to see how close we can get.

Bout 4 plan

Over the next bout, which is 5 weeks, our main focuses will be:

  • Hiring

    • In-person team: Generalist and specialist advisors for the in-person team.

    • Research team: A writer and researcher.

    • We don’t expect to finish the process in this time, but would like to have come up with a short list and have done at least two trials.

  • Improving online content for potential-100s.

    • Tech team: redesigning key web pages to better appeal to potential-100s - we hope to launch redrafts of several key pages.

    • Research: drafting the advanced series (aiming to finish at least one) and spare time into more podcasts or articles.

    • Niel: writing up the case for working on AI policy in the US govt.

Improving the online content is mainly a continuation of our focus from earlier in the year. In brief, we think we can likely make the site several times better at appealing to rated-100+ plan changes, as well as produce content that directly feeds into these plan changes (e.g. making the case for working in the US govt on AI policy, which we think is a top option, but isn’t widely pursued in our community). So, we decided to continue with this priority.

The decision to focus on hiring was more complex. Some of the key considerations included:

  • We think hiring is likely the best way to grow the impact of the in-person team at the margin and make our growth targets. This is because we have far more great candidates we are unable to speak to (~2000 on our waitlist, and ~20 new “tier 1” applicants per month), so there are few diminishing returns if we hire 1-2 people now.

  • Although we still think we can increase the efficiency of the coaching a great deal, we need to maintain a sufficient baseline of coaching in order to measure this. It was tough to do product design and maintain a high enough baseline, so it makes sense to hire 1-2 more people.

  • “Bus” risk - there is only one person on the research team right now, which would place us in a tricky position if they left (or were hit by a bus). It seems worth training up at least one extra person on each main function, to make us more robust. Staff capacity is also likely our biggest long-term bottleneck, so it make sense to take the low hanging fruit on hiring each year.

  • Timing. It’s useful to sync up with Open Phil’s process, since they are after a similar skillset. It’s also useful to sync up hiring within 80k.

Other key progress in the last 3 weeks

We finished planning and got started on the bout priorities early. This included:

  • We started hiring e.g. wrote up the process, wrote role descriptions, started contacting referrers and started speaking to leads.

  • Released a podcast with James Snowden about GiveWell’s cost-effectiveness estimates and Tanya Singh about operations roles.

  • Started redesigning the problem profiles page to be a better landing page, because it has become one of our top 5 most viewed pages.

We also pushed ahead with the office (e.g. viewed 5 more options), took some holiday, did some small miscellaneous tasks, and dealt with a major data breach of Typeform.

Have a great week,


Reply all
Reply to author
0 new messages