Development Livestreams

Do you dev livestreams?

I hope you do, so far it's done wonders for my main development project, Sara the Shieldmage, and for my most recent completed project, The Hero's Journey. It turns out there is an audience who are gamers, but also enjoy learning about the process of making games.

In the two months that I've been doing weekly dev livestreams, it's been interesting, because it sort of works as a dual marketing and development channel. It's marketing in that my games are always getting a little bit of traffic on the day after my streams, which helps me. But, it's also development in that I can just go about my usual workflow, but ask the audience their opinion on a decision point that would normally be just a blind guess.

What are your thoughts on Dev Livestreams, and in your opinion, how can we use them to more efficiently bring players into our development process...while still building hype?

 

@Asvarduil,

Can you tell us more? The idea of just streaming the entire time I'm developing ... it seems odd and I'd like to know a bit more about your setup, what you stream, for how long, how people find it, etc...

Gigi

 

Well, for starters, it's easier to show you.

As of this writing, this is my most recent livestream:

Generally, it is exactly what you said - I do some development on camera with a live studio audience*.

There is some necessary preparation, though, as this is a presentation to my stakeholders:

1. I select a generally-convenient time for my stream, because I want as many people in on this as possible.
2. I always lead into my stream by talking about what I've done during the week, not unlike a Scrum.
   
3. I try to walk into the dev session with 1:30 of material, spanning two subjects. (So, 0:45 each. I try to avoid going over 0:50 of material, because that's the point at which the adult mind stops optimally absorbing information.)
4. I try to select subjects that help people write their own games. There are many reasons people watch stuff, but the two big ones are that either something is inherently interesting, or something helps you do something you want to do.
5. I try to select subjects that are interesting. More about my experiments on this exists below.
6. I try to select subjects that I have two competing ideas about what to do with, that I can give to my audience.

People find the stream, because I tell them about it. I already have a few YouTube followers, but I also post in the Anook Community as there's some interest in game design from the 'core' gamers.

With selecting subjects, I've been doing some experiments on my streamers. Generally, easily-exposable things are better subjects than mentally-intensive things. I've noticed significantly improved audience retention in episodes where I do visual art and music composition, compared to story-writing (control subject, as it's intentionally boring), level design (this one struck me as odd), and programming (though people want to see programming on-screen, it's not really riveting viewing.)

There is minimal processing that goes into these streams. Some of that is my own hardware limitations. More of it, is because the point is for clear, honest, and unfiltered communication with the stakeholder; sure a second screen that I use to throw up things like background music or effects could help, but anything additional would (correctly) come off as making it seem more like an out-and-out marketing ploy than a development session (it's both.)

Also the rules are to be broken. In the linked stream, I go through about four topics, because the two I had selected were too short. While I personally felt this devalued my presentation, it actually seemed to be good for my audience because I was able to ask them, 'OK...what do you all want to see?' I was able to build some investment, which has its own overall value.

The ethics of this sort of dev stream are simply, "to communicate." I am not selling my game. I am not selling 'the game developer lifestyle.' I am showing a nascent game, and the processes going into it. Even though the point is to encourage interest (thus...'selling my game'), this is never a focal point of a broadcast, ever. If I broadcast right, this is a natural, and happily-received outcome.

Gigiwoo, does this answer some of your questions?

*: Just replace 'studio' with 'remote', and it's accurate.

 

i thought about doing this to bring awareness to my game pretty cool

 

The idea is not mine, actually. The predecessor I got it from, was the team behind Final Fantasy XIV: A Realm Reborn. Part of their resurrection of that game involved actually talking to users and understanding what gamers want.

http://www.gdcvault.com/play/1020797/Behind-the-Realm

 

oh ok ive seen it before in the last year or so...i didnt become interested until recently i was going to go the twitch route tho and then upload to youtube. I subscribed to your channel

 

Well, dang! That's cool. Helps you in lots of ways, and maybe, over time, helps some young developers too. Kudos.

Gigi

 

We used to do dev livestreams on a weekly basis about 1.5 years ago... They're easy to do and fun, but they didn't turn out very well for a variety of reasons.

The main problem was that livestreaming can't be the primary means of communicating with the community. Very few people have the time/conditions to sit in on an hour-long stream, and even fewer will want to watch the video afterwards. That leaves you with a very small subset of the community which took part in the stream and got some really in-depth information, and a vast majority which will have to hear about it second-hand.

That means streaming is only really effective for letting the dedicated enthusiasts have an insider look at the development process, but any announcements have to be made in a more concise format, especially if you're announcing plans for upcoming features/updates.

The second problem isn't unique to livestreams, but livestreaming is particularly affected by it. The problem is that with the pressure to do regular casts, you tend to speak because you need to say something, not because you have something to say. This can lead, at best, to uninteresting sessions, or at worst, saying stuff you didn't want to say, live, and on record.

We still do livestreaming for KSP, but we very rarely have devs on anymore, and when we do it's never to say anything important. Streaming is a good way to have fun with the community, play the game, do a show-and-tell of how something works, but it's a really bad way to communicate anything important.

Cheers

 

We've been streaming Sunday Dev Hangouts for several months now as an experiment, and use it as a channel to interact with our most active community members. It's not for fellow devs, because that isn't our audience. We share the very latest builds for a look at what's being worked on for the next monthly patch, as well as community education in how to get the most from the Editor (since we have community generated content). Those active members usually go on to answer newcomer posts in the forums, saving us time as a small team of 9.

We've found that a lot more folks (relatively speaking) watch the archived videos of the hangouts over on YouTube instead of turning up for the stream. But the figures are low (circa 500) and likely misleading, especially since we have far more YouTube subscribers than Twitch followers.

I wouldn't stream as a developer if revenue generation was the objective. It's about having a direct channel and engaging with our biggest advocates, and transparency (especially important if you're in Early Access given some of the bad press it's had of late). We get lots of great ideas and questions that help improve the game, and the conversations often continue over on the forum.

I agree with HarvesteR about streaming not being a good choice for announcements (Steam announcements are best in our experience). Our monthly dev blogs are however one of our primary means of communicating key information to the broader community, and we often keep content back so there's always a surprise even for stream followers. The dev blogs receive 1000% more views than the hangout streams, although that's because we announce them on Steam. They also take 2-3 days to make, which is significantly more prep than is required for 4 streams of 2 hours each month.

Whether streaming is worth it or not is more of a qualitative metric for us. As harvesteR said, it's more about being a fun thing to do with the community.