What’s next for the Pop Polls?

So, we’re almost at the end of the Modern Qualifiers on the new People’s Pop site. It felt like a good time to take stock of what’s working well and what isn’t in my efforts to move the PPPolls off Twitter and onto something else.

I’m going to get into the good points and the problems in separate bits, but there are two overriding things that kind of inform everything else and will ultimately dictate what happens to the polls next.

The first is that I really enjoy doing the polls. I love the community and the way it’s been an engine for discovery and fun. Ultimately that’s more important to me than how or where they happen.

The second is that it’s taking me vastly longer to do them.

It’s slightly hard to measure this, because I spent a fair bit of time on the polls before when I wasn’t strictly working on them but was refreshing, enjoying the chat, checking in, fretting, etc. And there’s stuff that I’m doing upfront now before a poll goes up – sourcing sleeves etc – which I was already doing as part of an “origins” post. Doing it as part of prep feels more laborious – cos I’m not getting the dopamine hits of responses/comments, I guess – but actually it’s not taking any more time.

But even taking those things into account, a much more significant chunk of my day is being taken up by poll work. I’m not 100% sure WHY it’s so much more work, especially given the raw number of matches is lower each day, but it is. And if I’m honest, it’s an unsustainable level of work, especially given that Life Developments elsewhere means I might have less time anyway.

So those are the basic issues. I don’t want to stop doing the polls. I need to stop spending this much time on them.

Now onto the move itself, and what’s working well. Some things are working both well AND badly, or at least very differently.

WHAT’S WORKING

First up, the site itself. I love it! I love how good the posts are looking, how well integrated the sleeves and playlists are in the posts. Alan and Steve have done an absolutely fantastic job of building a polling site that works and integrating it with the other things I do. It gives us all a great resource for doing music based polling and building a community around it.

The increased use of Curated Groups has been fantastic and something which should continue well beyond the Charity poll IMO.

I like the flexibility the plug-in gives me to extend matches until they get a certain number of votes, or to keep tied matches open to get a definite result.

I really like the possibilities the plug-in offers to do things which break away from the standard issue group-of-four format, even though I’ve not done that yet.

I think it’s great that the comments allow longer-form commentary and that I can keep substantial posts sticky.

On a personal level, I’m spending less time on my own Twitter, which is good, as Twitter isn’t that healthy for me.

We’ve got a solid core of participants and I hugely appreciate their efforts to make this strange new version of the People’s Pop Poll work.

The issue of hives, bots, etc is no longer a problem.

And finally, I wanted to at least find out how the polls work away from Twitter. Whatever I decide happens next, I’m deciding it from a base of actual evidence. It’s always good to try stuff! 

WHAT’S NOT WORKING

Obviously the big issue here for me personally is the time and work one, outlined above. But assume that magically I was able to spend half the time I do on the polls. Are they working away from Twitter?

The short answer is: not brilliantly, in the format we’ve been used to. There are things Twitter enables which turn out to be extremely difficult to replace.

As I’ve always said there are two points to the polls – discovery and fun. Discovery – people get to find new songs. Fun – we get to see together which songs win and argue about it.

Both of these rely on engagement, and so the level of engagement matters.

Discovery is holding up OK – we have a terrific crop of songs and there are a lot of new favourites to find. I haven’t worked out how best to run Golden Beats in the new format but it’s a solvable problem. 

Obviously the more people who take part the more likely it is your special song will find someone who loves it, but I think most (say 80%) of the people who were using the polls for discovery are still on board, if you take “average voting level in a Golden Beat match” as a yardstick.

(That aside, the level of overall participation is way less than it was on Twitter, obviously, and I was expecting that. I’ll be honest, it’s lower than I’d hoped by about ?, but also this is a 21st century poll with a lot of obscurities, so I should probably have calibrated a little lower! This is a platform popularity issue – it would be similar on eg Mastodon I think even if there wouldn’t be the format shock.)

Discovery is only one half of the polls, though – the fun half is a lot trickier. I’m not using “fun” just in the sense of “is this an enjoyable thing to be doing”, I’m specifically talking about the game part.

I think the biggest casualty of the move, personally, is tension. Close matches don’t feel close. Twitter brought a lot of people, but the real benefit it had is that it brought them together at the same time. There’s no urgency to the new site, no real attempts to shift close matches, and no discussion about the actual OUTCOMES of the matches, just some conversation about the music itself.

Now, conversation about the music itself is the heart of what I want to do! Without it you end up like Music Twitter in general, just endlessly listing the same LPs like you’re rearranging the shelves of your garden shed. 

But the tension is what makes the polls feel like a game, or an event, and what encourages people to bring new participants in. The game is a skeleton to hang music discovery on, but t’s a very necessary skeleton. And the move has taught me how much that mattered in terms of my own motivation and keeping the community going.

The other big issue – which is tied up with this – is a format one.

The format of the polls is based on the limitations of Twitter polls – four options at a time, one choice only, a set duration. We refined that into a structure that worked very well.

The Charity Crusher is locked into that structure because of the Curated Groups, but it’s very ill suited to the blog format, which continually pushes content down and makes it harder to recover and boost. Twitter gives you a lot more control of the “now” of the poll – you can shift focus easily onto close matches, admin, etc. Which is where a lot of the tension and community engagement comes in.

A lot of the issues come from trying to recreate that format on the blog instead of trying to do something new.

Finally, the move away from Twitter is a bit of a fake so far. I have to promote new polls on there to get any traction – 70% of visits are via Twitter. (This promotional work adds to the sense of extra labour). On some level what we’ve been doing is only really possible using Twitter (directly or indirectly), and we wouldn’t be the only fun activity to find that this is the case. I hope it’s still possible to build a community off Twitter, but it always took a long time even in the web 1.0 era.

SO WHAT HAPPENS NEXT

I have several problems to solve.

  1. My poll workload needs to be lower
  2. The polls and discussion aren’t as fun (in a game sense)
  3. We’re trying to run a Twitter formatted poll away from Twitter.
  4. The polls are dependent on Twitter for the near to medium future.

But we also have the considerations that led us here in the first place, and some new opportunities.

  1. Twitter may work well for music polling (assuming polling isn’t deliberately broken at some point) but it’s a bad place owned by a guy with terrible intentions. Even if moving away permanently is much harder than I hoped, I don’t want to be tied to it.
  2. We have a sleek website designed for music conversation and polling to use as a foundation which can do a lot more than just groups of 4 Twitter polls (but which isn’t good for synchronous conversation).

We’re about ? of the way through the Modern poll, with the Ancient poll to come and two excellent topics (2015 and Late Work) on the slate.

Here’s what I think are the options. The first is more or less the status quo. Not keen on this.

  1. KEEP GOING: Run the Modern and Ancient polls on the new site, then work out how best to do the 2015 and Late Work ones. Solve the workload problem by simply doing it all slower (2-3 polls daily, not 5). Accept the shift in discussion away from “game+music” to just “music”.

The next two are ‘nuclear options’ – basically saying, the move was a mistake, scrap it, and either carrying on or finishing the polls. Even less keen on these.

  1. PERMANENT TWITTER RETURN: Basically go back to where we were earlier this year and if Twitter dies we call it a day.
  2. LET IT GO: Call it a day anyway but in a controlled way. Run the Ancient and Modern poll then move into the all-Pollhalla final tournament.

The next one is basically back to where we were in January. Not keen on this either but any solution which acknowledges a need for Twitter probably involves at least looking at Mastodon or Bluesky too.

  1. SECRET THIRD THING: Look for an alternative platform (Mastodon, Bluesky, Reddit) for Twitter-style polling once the A&M polls are done.

And finally the ‘retrenchment’ options. Run the current polls on Twitter, and in the meantime work out what format takes advantage of the new site, and either switch the polls to it or run them in parallel with it.

  1. TEMPORARY TWITTER RETURN: Run the Ancient poll (and potentially finish the Modern one) on Twitter, where the format works. Meanwhile work out a more blog-friendly format which would suit future themed nominations-based events and move back to the new site to implement it for the 2015 poll.
  2. HYBRID/BEST OF BOTH WORLDS: Return to Twitter for tournament-style polling at a smaller scale and simultaneously use the PP website to do discovery and fun games more suited to it (eg “league” rather than “tournament” games, LP-based polls, slower all-discovery polls, etc.)

All the options which involve Twitter would mean less polling than we did before – probably a permanent shift down from 6-8 daily polls to 4-5.

What do I think? My gut instinct says 6 – the hybrid approach, keeping the old spirit intact while building something new with the new architecture. But I’m also aware this feels like having my cake and eating it. I don’t want to be really hasty about it, and I’d love to know what other people think too and what options I might not have considered!

Finally, thanks again to Alan and Steve for building something awesome which gives us these possibilities, and to the people who’ve come over (or returned from exile) and valiantly helped the Pop Polls survive outside their natural habitat!