Hey Cillian — thanks so much for a really thoughtful/detailed question! 

I'll take this one since I was the only staff member on marketing last year :)

The short answer is: 

• Marketing ramped up considerably over the second half of 2022. Web engagement time grew a lot more in the second half of 2022 as well — if we just compare Q3 &Q4 2020 and 2022, engagement time grew 50% (rather than 10%).
• But that still doesn't look like web engagement time rising precisely in step with marketing investment, as you point out!
  • We don't know all the reasons why, but the reasons you gave might be part of it.
  • There are five other reasons I want to mention: 
    • marketing didn't focus on engagement time
    • people we find via marketing seem less inclined to use our advice than other users
    • there’s a headwind on engagement time
    • site engagement time doesn’t count the book giveaway
    • 2020 was a bumper year!

The long answer...

• It's true that there's only a 10% rise in engagement time between 2020 and 2022.
• The main thing going on here is that marketing investment didn't rise until the second half of 2022 — I spent a while trying out smaller scale pilots. So the spend is very unevenly distributed.
• If we compare just Q3 and Q4 (i.e. Q3 and Q4 2020 with Q3 and Q4 2021 and Q3 and Q4 2022), there was a 6% fall from 2020 to 2021, and then a 59% rise from 2021 to 2022, resulting in an overall 50% rise from 2020 to 2022!

But our marketing budget still rose by a lot more than 50% — so what's going on there?

I'll start out with the reasons you gave then add my own:

• Maybe the price of acquiring new users / engagement hours increases geometrically or something

Yeah I don't exactly know how the price of finding new users increases, but we should probably expect some diminishing returns from increased investment. 

• It looks like marketing drove a large increase in newsletter subs. Maybe they're engaging with the content directly in their inbox instead?

I think this might be a smallish part of it — I've noticed an effect where if the email we send on the newsletter is itself full of content, people click through to the website less than if the newsletter doesn’t itself provide much value.

I don't think this can account for tons of what we're seeing, though, just cos I don't think the emails work as a 1:1 replacement for 80k's site (I can't really imagine there being much of a 'substitution effect' here).

• Maybe you expect a lag in time between initial reach & time spent on the 80k website for some reason (e.g. because people become more receptive to the ideas on 80k's website over time, especially if they're receiving regular emails with some info about it)

I think this might be a pretty big part of what's going on. 

There does seem to be a significant 'lag time' from people first hearing about us and people making an important change to their careers (about 2 years on average) and I think there's often a lag before people get really engaged with site content, too.

Also, bear in mind that because of what I said about the 'unevenness' of growth from marketing, people who found out about us this year are mostly still really new.

• Maybe marketing mainly promoted podcast / 1-1 service / job board (or people reached by marketing efforts mainly converted to users of these services)

Yep, I did put some resources directly towards promoting the podcast, and a much smaller amount of resources towards directly promoting 1-1 (about 7% of the budget as a whole, and probably more like 10% of my time). So this could be (a small) part of what's going on.

There are five other main things I think explain this effect:

• Marketing didn't focus on engagement time
  • My focus was really on reaching new users, rather than trying to get people to spend time with our content. 
  • For example, last year I spent almost zero effort and resources on trying to get people who already knew about 80,000 Hours to engage with stuff on the site. 
• People who find out about us via marketing are, on average, less interested in our advice
  • I talked about this a little bit in my post that you linked. 
  • I think this is probably a pretty big factor here!
• There's a headwind on engagement time, i.e. engagement time by default seems to go down over time
  • We think this is because, e.g.:
    • Articles that were last updated longer ago are seen as less relevant by search engines and deprioritised
    • Things we wrote become out of date and less useful to people
    • There might be some broader internet trend where people are spending more time on social media and less time on individual websites
• Our current metric doesn't incorporate engagement time from reading the books in the book giveaway
  • However, we are looking into including it in future reporting!
  • My (still ongoing) analysis of survey results about the book giveaway suggest that this engagement time is at least tens of thousands of hours, possibly >100k hrs. 
  • (For context, our average monthly engagement time since Jan 2021 is about 8,000 hours (ha!)).
• 2020 was weird
  • We saw a very large spike in traffic in early 2020 — partly because we had content on COVID when many places didn’t, and partly because of the broader trend where tons of sites got a lot more traffic as people were staying home and spending more time online.
  • So 2020 might be an “inappropriate benchmark” or something like that.

Okay I hope that gives you an insight into what I think is going on here! Sorry for length :)

Hey! Arden here, also from 80,000 Hours. I think I can add a few things here on top of what Bella said, speaking more to the web content side of the question:

(These are additional to the 'there's a headwind on engagement time' part of Bella's answer above – though they less important I think compared to the points Bella already mentioned about a 'covid spike' in engagement time in 2020 and marketing not getting going strong until the latter half of 2022 .)

1. The career guide (https://80000hours.org/career-guide/) was very popular. In 2019 we deprioritised it in favour of a new 'key ideas series' that we thought would be more appealing to our target audience and more accurate in some ways, and stopped updating the career guide and put notes on its pages saying it was out of date. Engagement time on those pages fell dramatically.

   This is relevant because, though engagement time fell due to this in 2019, a 'covid spike' in 2020 (especially due to this article, which went viral: https://80000hours.org/2020/04/good-news-about-covid-19/) masked the effect; when the spike ended we had less baseline core content engagement time.

   (As a side note: we figured that engagement time would fall some amount when we decided to switch away from the career guide: ultimately our aim is to help solve pressing global problems, and engagement time is only a very rough proxy for that. We decided it'd be worth taking some hit in popularity to have content that we thought would be more impactful for the people who read it. That said, we're actually not sure it was the right call! In particular, our user survey suggests the career guide has been very useful for getting people started in their career journeys, so we now think we may have underrated it. We are actually thinking about bringing an updated version of the career guide back this year.)

2. 2021 was also a low year for us in terms of releasing new written content. The most important reason for this was having less staff. In particular, Rob Wiblin (who wrote our two most popular pieces in 2020) moved to the podcast full time at the start of the year, we lost two other senior staff members, and we were setting up a new team with a first time manager (me!). As a result I spent most of the year on team building (hiring and building systems), and we didn't get as much new or updated content out as normal, and definitely nothing with as much engagement time as some of the pieces from the previous year.

3. The website's total year-on-year engagement time has historically been greater than the other programmes, largely because it's the oldest programme. So it's harder to move its total engagement time in terms of %.

   Also, re the relationship of these figures with marketing, the amount of engagement time with the site due to marketing did go up dramatically over the past 2 years (I'm unsure of the exact figure, but it's many-fold), because it was very low before that. We didn't even have a marketing team before Jan 2022! Though we did do a small amount of advertising, almost all our site engagement time before then was from organic traffic, social media promotions of pieces, word of mouth, etc.

(Noticing I'm not helping with the 'length of answer to your question' issue here, but thought it might be helpful :) )

Thanks Vaidehi -- agree! I think another key part of why it's been useful is that it's just really readable/interesting -- even for people who aren't already invested in the ideas.

Basically: these just take a really long time!

Lumping 2021 and 2022 progress together into a single public report meant that we saved hundreds of hours of staff time. 

A few other things that might be worth mentioning:

• I’m not sure whether we’ll use 1 or 2 year cycles for public annual reviews in future. 
• This review (14 pages + appendices) was much less in-depth and so much less expensive to produce than 2020 (42 pages + appendices) or 2019 (109 pages + appendices). If we end up thinking that our public reviews should be more like this going forward then the annual approach would be much less costly. 
• In 2021, we only did a ‘mini-annual review’ internally, in which we attempted to keep the time cost of the review relatively low and not open up major strategic questions. 
• We didn’t fundraise in 2021. 
• I regret not publishing a blog post at the time stating this decision. 

Hi Vaidehi - I'm answering here as I was responsible for 80k’s impact evaluation until late last year. 
 

My understanding is that plan changes (previously IASPC's then DIPY's) were a core metric 80K used in previous years to evaluate impact. It seems that there has been a shift to a new metric - CBPC's (see below). 

This understanding is a little off. Instead, it’s that in 2019 we decided to switch from IASPCs to DIPYs and CBPCs. 

The best place to read about the transition is the mistakes page here, and I think the best places to read detail on how these metrics work is the 2019 review for DIPYs and the 2020 review for CBPCs. (There’s a 2015 blog post on IASPCs.)

~~~

Some more general comments on how I think about this:

A natural way to think about 80k’s impact is as a funnel which culminates in a single metric which we can relate to as a for profit does to revenue. 

I haven’t been able to create a metric which is overall strong enough to make me want to rely on it like that. 

The closest I’ve come is the DIPY, but it’s got major problems: 

1. Lags by years. 
2. Takes hundreds of hours to put together. 
3. Requires a bunch of judgement calls - these are hard for people without context to assess and have fairly low inter-rater reliability (between people, but also the same people over time). 
4. Most (not all) of them come from case studies where people are asked questions directly by 80,000 Hours staff. That introduces some sources of error, including from social-desirability bias. 
5. The case studies it’s based on can’t be shared publicly. 
6. Captures a small fraction of our impact. 
7. Doesn’t capture externalities. 

(There’s a bit more discussion on impact eval complexities in the 2019 annual review.)

So, rather than thinking in terms of a single metric to optimise, when I think about 80k’s impact and strategy I consider several sources of information and attempt to weigh each of them appropriately given their strengths and weaknesses. 

The major ones are listed in the full 2022 annual review, which I’ll copy out here: 
 

1. Open Philanthropy EA/LT survey. 
2. EA Survey responses. 
3. The 80,000 Hours user survey. A summary of the 2022 user survey is linked in the appendix. 
4. Our in-depth case study analyses, which produce our top plan changes (last analysed in 2020).  EDIT: this process produces the DIPYs as well. I've made a note of this in the public annual review - apologies, doing this earlier might have prevented you getting the impression that we retired them. 
5. Our own data about how users interact with our services (e.g. our historical metrics linked in the appendix). 
6. Our and others' impressions of the quality of our visible output. 

~~~

On your specific questions: 

• I understand that we didn’t make predictions about CBPCs in 2021. 
• Otherwise, I think the above is probably the best general answer to give to most of these - but lmk if you have follow ups :) 

Hi — thanks for for the question! That’s definitely what we care about most, but it’s also unsurprisingly very hard to track, as you say. We have different ways we try to assess our impact along these lines, but the best metrics we can share publicly are in an appendix to our two-year review that summarises the results of our user survey. You can also see Brenton's answer in a separate comment for much more detail about our efforts to track these metrics.

Thanks for the thought!

You might be interested in the analysis we did in 2020. To pull out the phrase that I think most closely captures what you’re after: 
 

We also attempted an overall estimate. 

This gave the following picture for 2018–2019 (all figures in estimated DIPY per FTE and not robust):

1. Website (6.5)
2. Podcast (4.1) and advising (3.8)
3. Job board (2.9) and headhunting (2.5)

~~~
 

We did a scrappy internal update to our above 2020 analysis, but haven’t prioritised cleaning it up / coming to agreements internally and presenting it externally. (We think that cost effectiveness per FTE has reduced, as we say in the review, but are not sure how much.)
 

The basic reasoning for that is:

1. We’ve found that these analyses aren’t especially valuable for people who are making decisions about whether to invest their resources in 80k (most importantly - donors and staff (or potential staff)). These groups tend to either a) give these analyses little weight in their decision making, or b) prefer to engage with the source material directly and run their own analysis, so that they’re able to understand and trust the conclusions more. 
2. The updates since 2020 don’t seem dramatic to me: we haven’t repeated the case studies/DIPY analysis, Open Phil hasn’t repeated their survey, the user survey figures didn’t update me substantially, and the results of the EA survey have seemed similar to previous years. (I’m bracketing out marketing here, which I do think seems significantly less cost effective per dollar, though not necessarily per FTE.)
3. Coming to figures that we’re happy to stand behind here takes a long time. 

~~~

This appendix of the 2022 review might also be worth looking at - it shows FTEs and a sample of lead metrics for each programme.