Featured Blog | This community-written post highlights the best of what the game industry has to offer. Read more like it on the Game Developer Blogs.
Do betas & demos right, the Monster Train way! 2
Are you using demos solely as a marketing tool, or are you actually engaging your community, iterating, and improving the game using demos or betas? If the latter, we’re way more of a fan. Which is where Monster Train comes in...
[The GameDiscoverCo game discovery newsletter is written by ‘how people find your game’ expert & GameDiscoverCo founder Simon Carless, and is a regular look at how people discover and buy video games in the 2020s.]
Welcome to a brand new game discovery week! As we kick things off, this lead story comes from an interesting chat about game demos that we had on the GameDiscoverCo Plus-exclusive Discord at the weekend.
How should you approach game demos (& betas)?
Intro: we at GameDiscoverCo have perhaps been overly negative about demos recently, and I won’t rant extensively. But our key belief, for the record: demos alone aren’t much of a marketing win, since so many games have them nowadays. You’re largely preaching to the converted. And you can even turn those ‘converted’ off, if you don’t get your demo right. (Also - it takes a chunk of dev time to make a truly good demo! And player conversion rates are likely surprisingly low.)
It’s very difficult to prove this belief, though! Demos tend to show ‘positive metrics’ (measurable downloads and wishlists) which feel good. And I agree there are situations where demos can expand reach more significantly, if deployed alongside platform-level promotions. Steam Festivals are one of them - I’d recommend everyone put a limited time demo in those, because of both player and streamer interest in those festivals.
But I realized a major question is this: are you using demos solely as a marketing tool, or are you actually engaging your community, iterating, and improving the game using demos or betas? If the latter, we’re way more of a fan. Which is where Monster Train comes in…
I’ve known Shiny Shoe’s Mark Cooke for a few years now. And before his company worked with Good Shepherd to have a smash hit (almost 10,000 Steam reviews, 98% positive!) with roguelike deck builder Monster Train, they were a well-regarded studio on titles like the Full Throttle and Day Of The Tentacle remasters. So they know what’s up.
Anyhow, Mark’s strategy for Monster Train’s beta and demo is pretty close to perfect for ‘showing playable versions of your GaaS-ish game to players before release’. And it’s centered around improving the game, but also works as marketing! So he kindly gave me permission to reprint his strategy, as discussed in our member-only Discord:
“Leading up to the Monster Train release, we ran: 1. A pre-release private beta for ~7 months which required an NDA. (It's still ongoing, we still have some fantastic testers in there helping test/balance pre-release content.)
2. A pre-release public beta for ~6 weeks which did not require an NDA but required people join our Discord to get a Steam key for the main Steam app ID. I think we ran this around 2-3 months before launch. It had limited content. We revoked all these keys after the beta ended. We made sure to give this to a bunch of streamers right before it started (so there was some exclusivity), which really helped.
3. A public pre-release demo [for 3-4 weeks immediately before launch] using the normal Steam demo flow. It was basically the same thing as the pre-release public beta, but with bug fixes and polish from the data/feedback we got in the public beta. We later removed this from Steam when the full game launched.”
There’s obviously some possible variants here. (You could do a public demo in a Steam Festival instead, and you could use Steam’s new Playtest feature - though that doesn’t get them onto your Discord.) But in general, this multi-phase approach is on the money.
Mark also talked about a particularly interesting (and advanced) feature Monster Train had: “One incredibly valuable thing from the public beta / demo was we had a UI element on screen at all times that said "press F8 to send us feedback." It opened a simple UI to type in your thoughts and automatically captured a screenshot, the player's log file, and the player's save game.
Having this call to action got us TONS of valuable feedback and bug reports. We fixed all of the common complaints before the game came out, which both made the game better, and gave us community talking points on developer Twitch streams leading up to release.”
As another example, No More Robots & PanicBarn had also done some survey-based player feedback for a Not Tonight beta which was run in a similar way (join the Discord to get the Steam key.) This allowed for both improving the game and positive PR/marketing, like so:
Mark says his team also ran a survey after the public beta and it was somewhat useful, but the in-game feedback was far better. He noted: “I think the main reason is when someone wants to complain about something that is on their mind, they are much more likely to tell it to you right when it pops into their head instead of at some later point. So it gets you a lot closer to the real pain points.”
Anyhow, all of this is hopefully food for thought. The general point: rather than just putting a generic demo or prologue up for months, use one or multiple limited-time betas or demos to get feedback, improve the game and increase hype. We’re convinced it’s the way to go. (And post-release demos: just say no!)
The game discovery news round-up..
And that’s a great start! Now, let’s move on to a look at some of the other game discovery and platform news that’s popped up over the past few days. And let’s start with some great math on wishlists and conversions:
Kitfox Games published a blog post from Systemchalk which analyzes the estimated value of a wishlist, even after release, for multiple Kitfox games - see above graph. It’s a super interesting and complex calculation which also takes into account the game being in X%-off sales. (Conversion percentages are fairly stable, but sale discounts get bigger, hence the decrease over time!)