2018 in review

I write this blog as the sun sets on 2018 out my window, and finding myself in a contemplative mood. So I thought it would reflect on my first year as a full time science communicator (well, I wasn’t creating full time for a fair chunk of the year, but I digress...). Because I’m me, I made a video at the start of the year detailing what I wanted to do in 2018:

To briefly summarise, I wanted to:

  • Vlog every week

  • Make a science video every week

  • Blog/write frequently

  • Be presenting/public speaking

  • Maybe do some livestreaming

In this blog post I’m going to assess how well I met these targets, what worked on my channel, what didn’t, and what my plans are looking forward to 2019.

What worked on my channel in 2018

Google analytics provides a wealth of information including video views, total watch time, subscribers gained, and much more. For the sake of simplicity, I’m going to use views as my metric in this simple evaluation, though it’s worth noting that watch time is more important to the YouTube algorithm. It also, however, makes a comparison between different kinds of content that I make difficult. So, using views as my metric, the ten most successful videos on my channel this year were:

  1. Could planets from Star Wars really exist? (SCIENCE)

  2. Why you shouldn’t apply for a PhD (PhD)

  3. Why sin and cos don’t mean anything (SCIENCE)

  4. Want to study physics? Read these 10 books (BOOKS)

  5. Why has it been so hot?? (SCIENCE)

  6. A day in the life of an Oxford physics student (OXFORD)

  7. From primary school to PhD (PhD)

  8. Climate change is already irreversible (SCIENCE)

  9. What I’m doing after my PhD (PhD)

  10. Handing in my PhD thesis (PhD)

To break down the list, there are four science/maths videos, four videos based on my PhD, one book video, and one video about my time at Oxford. This is not a surprise to me, as the thing I’m currently known for online is my PhD vlogs. A similar analysis for 2017 shows nine of my top ten performing videos were based on my PhD (the tenth, the number one spot, being A day in the life of an Oxford physics student). This encapsulates the biggest challenge I’ve faced this year - getting people to watch things I make even after I’ve finished my time at university. As such the fact that five of my top-performing videos this year were based on science, maths, or books is encouraging to me, though there are definitely significant challenges ahead.

An alternative metric to views, and a more practical one in terms of keeping the lights on, is the money earned by my videos. Due to the nature of how YouTubers fund themselves, there’s a binary classification system for my videos: those with Google adsense enabled, and those with sponsor slots. I don’t enable ads on sponsored videos for the first month after upload, as part of my contract with the sponsors. So the fee I get from sponsorships - tied to performance - is the lion’s share of revenue generated by the video. By this financial metric, the most successful videos on my channel were Why sin and cos don’t mean anything, Why has it been so hot??, Climate change is already irreversible, and Could planets from Star Wars really exist?, along with two other videos which did not perform as well in terms of views: Economics PhD students predict the next big thing, and How to build your own Mars rover. Interestingly the most profitable videos were all science-based, which lends me more optimism going in to 2019. Partly because I have demonstrated that I can make (partially) viral science content separate from my PhD, and so be paid a decent fee by sponsors, but also because some of these videos have had long ‘tails’, earning substantial money through Google adsense a long time after release.

What didn’t work on my channel in 2018

A simple analysis for under-performing videos released this year isn’t really possible using Google analytics, as the lowest performing videos by any metric will all be from eight years ago that nobody is watching any more (with good reason). However based on my personal experiences this year, I believe that the worst performing videos by views have all been video essays in the second half of the year, including The physics of mythical flying beasts, How to build your own Mars rover, and How science invented superman vision, along with the general performance of my regular vlogs.

There are various reasons why this has been the case, many of which are associated with how I’ve marketed these videos in both title and thumbnail. The under-performance of my vlogs has already been covered in a separate blog post. I am confident that the science videos I’ve made this year have only got better and better, and my writing and editing skills have definitely improved greatly. As such I’m happy with the quality of what I’ve been making. But I think that what my videos have lacked is 1) effective thumbnail/title combinations, and 2) a unique style that separates them from the pack of other science videos.

How well did I achieve my targets?

Let’s break these down one by one shall we?

Vlog every week

  • Er, lol, that didn’t happen

  • As detailed in a previous blog post, the regular video blog just wasn’t feasible any more so I pulled the plug on the series

  • I plan on vlogging irregularly in 2019, when interesting things happen in my life

Make a science video every week

  • This was also not achieved, but the miss was less severe

  • In 2018 I released twelve science videos, some of which were immensely popular and some were absolute flops

  • I think in this particular goal I was way too ambitious, and underestimated how much time goes in to making a good video essay. I also underestimated how much of the year I would be away from YouTube doing corrections to my thesis or generally recovering from finishing my PhD.

Blog/write frequently

  • Achieved!

  • I released six blogs, including this one, this year, and have been writing full scripts for almost all my videos

  • I’ve also been doing some creative writing on the side, have written full goodreads reviews for everything I’ve read, and committed more time to making book videos

  • I’m happy with how this part of the job has gone this year!

Be presenting/public speaking

  • Mostly achieved

  • This year I gave a variety of talks, including a TEDx talk, sessions at schools, universities and conferences both in the UK and abroad, on subjects ranging from video production to my PhD research

  • I did not however do any presenting work on contract, so this target was only partly achieved

Maybe do some livestreaming

  • Finally something which I smashed!

  • By the end of the year I made Twitch affiliate, streaming twice a week doing both science past papers and playing games, gaining nearly 2,000 followers

  • Together with closer associations with well known streamers like Hat Films and the Yogscast, this has been much more of a success than I’d dared to hope

Overall verdict: two and a half objectives achieved, two objectives mostly failed or altered due to circumstances. If I were to give myself a grade for this year, it would be a C. However I do think that my objectives were partly unrealistic, so I will need to modify them for the coming year.

What do I want to do in 2019?

As I just mentioned, one thing that I learned this year was to set achievable goals! I definitely think there’s merit in keeping an ‘impossible list’ to show how far you’ve come, but I also want to have actionable goals. This is particularly important as I derive a fairly substantial chunk of my income from my audience on Patreon. If you treat my channel as a company, and my patrons as shareholders, then those shareholders have a right to know where the CEO is taking the company!

Learning from my mistakes this year then, I will aim to:

  • Upload one video a week on my channel

  • Form a stronger visual/brand identity

  • Stream several times a week (most likely 2-3 times)

  • Increase the frequency of blog posts, and move towards writing original content appropriate for broader publication

  • Focus on science content relevant to atmospheric science and climate

  • Release ‘tentpole’ videos roughly every two months

  • Collaborate further with other YouTubers

  • Release a major, ambitious project

While much of the list speaks for itself, let me unpack the rest a little.

One of the great joys this year has been working with other creators, including Real Engineering, Hannah Witton, Hat Films, The Yogscast, and Tom Scott. I’d like to do more of that in the future please! Apart from expanding our respective audiences, collaborations are always a great deal of fun, and mean that I don’t just create content alone. Being a YouTuber can be very isolating, and so one of my objects this year is to ‘get out more’ and work further afield with others. This might take the form of individual videos, or possibly a more permanent, even physical, collaboration. I’ll leave that there for now.

By contrast, one of the biggest difficulties I’ve faced this year has been getting people to watch my videos! Of course, that’s the principal difficult of every creator, but I mean that in the sense that when people do watch my videos, particularly the science ones, the feedback has been great. What I’ve struggled to do this year is getting people to that point, by a combination of failures in titling and thumbnail design, but also by lacking a strong visual identity. One of my main objectives next year is to get my videos to stand out from the crowd, and make my channel a ‘must watch’ rather than a ‘put in my watch later for two months before unceremoniously deleting’.

On a more positive note, something I’ve noticed this year is that just one or two videos can have the hitting power of dozens put together, if the subject is chosen well and enough time is invested into their production. In particular I’m talking about Could planets from Star Wars really exist? and how this video has outperformed absolutely everything on my channel in the past few months. I refer to such videos as ‘tentpoles’ in that they can support the whole channel, keeping canvas over my head, if I can afford to invest time in their production. I tried to repeat the success of the Star Wars video with The physics of mythical flying beasts, but this was not successful despite being (I think) really rather good. So one of my objectives this year is to release a tentpole video every few months, investing time in a longer production cycle and planning ahead strategically. I very much hope that this will take the pressure off my regular uploads, allowing me to be a bit more experimental with my content while still being financially secure.

Then lastly I have a rather ambitious plan in store. I’ve not seen anyone attempt anything like it before. It stretches across several YouTube channels as well as its own dedicated website, and will involve multiple people. I’m hesitant to reveal any further details at this point, but I hope that 2019 will be the year of Planet Factory. Watch this space…


And that is that. If you stuck around til the end then good lord, you deserve a medal. Frankly, if you’ve put up with my content for a whole year then you deserve a medal anyway. To those of you who have supported me, watching my content and putting up with my mis-steps, encouraging me in the comments and on twitter, trusting in my dubious at best abilities: thank you. You made it possible for me to spend all of 2018 as entirely self-employed, allowing me to make what I think is important, and trying to make the world a better place. Hopefully this will be the first of many years that I will say this:

Thank you for the past year. Here’s to the next!