Intro

In a previous blog post, I introduced you to the Discovery Queue (DQ). DQ is a very important widget which is how top tier games get lots of traffic (Read that post here.)

In short, if you drive a TON of traffic from outside of Steam (like youtube, twitter, tiktok, reddit), the algorithm takes notice, and says 

“Whoa this kind of traffic correlates with games that sell a lot, so we should start treating this game as one that has potential.” 

  • The Steam Algorithm if it were a human being

Steam tells you why every game ends up in the DQ  right above the wishlist button: “Because it is popular” 

But what does “popular” mean? How do you get “popular?”

How does one become “Popular?”

I surveyed a bunch of developers who had been featured in the Discovery Queue to see what metrics they needed to hit to be featured. Here are the details of the games:

Games in the survey:

  • 14 games had significant Discovery Queue showings that were included in this study
  • All the data here is before the game was available for sale

Why the priority on pre-release games? In my experience, Steam gives even more DQ visibility once your game can be purchased. Basically every game gets a shot at DQ featuring at release. The numbers related to DQ-featuring upon release is a bigger topic and deserves its own blog post. So I will do that at some point.

But for this blog post, I am focusing just on pre-release DQ featuring because it is rarer, it is something you might be able to induce to get more wishlists before launch, and not all games get the opportunity. 

But first, entdeckungsliste

Before we dive into it thought, I couldn’t figure out where to put this so I will just tell you, the official Valve German translation that they put in the UI for Discovery Queue is entdeckungsliste. 

German is so awesome.

Let’s continue…

What it looks like when you get picked up by the Discovery Queue

If your game has never been blessed by the DQ fairy, let me explain what it looks like.

The following graph is from Steamworks and shows the total number of page visits and the total number of page visits that came from the Discovery Queue to a game.

Every DQ wave kind of looks like this.

There are 4 basic phases when you get picked up. See the graph and the labels below. This all typically lasts about 2 weeks.

A) The Inciting Incident. This is the big burst of external traffic from some big marketing promotion to your game. Notice how visits spikes at point A, but DQ hasn’t really woken up yet.

B) Peak DQ traffic. At point B the DQ woke up because the Steam algorithm saw the A event. As a result of A, the DQ floods your game’s page with traffic. Notice how immediately after A, the visits start to dip and then the DQ peaks (B) and the total visits skyrocket back up. This is usually a sign that a combination of all the Steam widgets are showing off your game (but mostly the DQ).

C) The long DQ tail. Although the DQ peaked a couple days earlier, the DQ continues to feature your games for days. This usually lasts 2 weeks (more on that in a section below).

D) The DQ dip. At some point, the party ends and the DQ doesn’t feature your game ask much. But it never really goes away. This dip can be a bit fuzzy as to when it really ends.

How many visits/wishlists does it take to trigger the DQ?

So, how much traffic, how many wishlists does it take to trigger the DQ? Basically, how big does A have to be before B wakes up?

To figure this out, I asked participants to find a DQ spike (B) and then go back 1 week BEFORE and report how many visits and wishlists they earned. Basically, how many wishlists do you need to generate to trigger the DQ?

This table shows the results. It presents how many Visits and Wishlists a game earned during the 7-day period before a major Peak in DQ Traffic aka (B) in the chart above.

VisitsWishlists
Average3455313840
Median112023583
Lowest number that still triggered the DQ4347524

To help visualize what is typical, I graphed all 14 games that had a significant enough DQ spike.

In the graph below, each blue bar represents a different game in my survey. The bar height is the number of views they got in the 7 days before DQ featuring. As you can see, although there were some outliers, but most of the time around 10,000 views got you on the DQ. 

The red line is the highest 1 day peak of the DQ. I was curious if having a higher wishlist count means a higher peak? There does not seem to be a correlation.

In the most recent Valve presentation about how the Steam widgets work, Valve indicated that they don’t really trust “visits” as a good measure of a game’s quality. I think because that can be gamed fairly easily with a bot farm. Instead, Valve uses wishlsits as a stronger indicator of whether a game is interesting to Steam customers.

So, lets look at the number of wishlists you need in a 7-day period to trigger the DQ. As you can see, in the graph below, that they all seem to group around the 1500 to 4000 wishlist range. Similar to visits, the DQ peak does not seem to correlate with the number of wishlists. 

How long will you spend in the DQ?

It is a magic feeling when the DQ favors your game. It is almost like surfing: you just wonder how long you will stay in the wave before it dumps you. In this question of the survey, I asked developers to send me how long it was from point B (the peak) to point D (the dip).

Here is a graph of each of the featured games and the time they spent in the DQ (note these are sorted by the number of days so the games do not align with the other graphs).

Basically, you get 2 weeks in the DQ. Enjoy the ride.

How do you get into the DQ?

Valve is a bit cagey about the specifics other than it is algorithmic and based on the games you like to play.

In my survey I asked developers to tell me what marketing actions they took to induce DQ featuring. There were 5 main responses

  1. Announcing their game / Launching a coming soon page
  2. Getting featured by a Streamer
  3. Getting a social media post to “Go Viral”
  4. Running a kickstarter
  5. A festival

Launching a Coming Soon Page

This was a huge cause of DQ featuring. Many developers said they launched a trailer, and shared it on social media. 

Actions you can take:

Treat the launch of your Steam Coming Soon page as a marketing beat. A “marketing beat” is just a term we use in the industry to throw a party for your game. So don’t just turn on your Steam Coming Soon and go on with the rest of the day. No no no. You need to build up to it. First, make a trailer. Trailers are your lottery ticket to coverage. Don’t just launch with a bunch of random screenshots. 

Trailers don’t always work, but you can’t win if you don’t play.

And when you have your trailer, don’t just dump it on youtube. Instead, spend weeks teasing your pending announcement. Count down to it. Reach out to press ahead of time to see if they are interested and want to feature it.. Then on announce day get everyone you know to share the trailer and the Steam page on social media. Make a cut down version of the trailer and post it on TikTok, Instagram Stories, Youtube Shorts. 

Put out your game announcement everywhere. Maybe it will catch on and go viral. That virality MIGHT trigger the DQ. 

Other examples of games that had big launches:

A streamer covering your game

We all know Streamers are important but many people don’t understand why. Yes the direct traffic from Youtube is good, but the secondary effect that makes it REALLY good is Steam often uses that traffic as a reason to put your game in the DQ.

A couple developers were

Actions you can take:

If your wishlists are flat and your game seems stuck in a rut earning fewer than 10 wishlists a day, get a demo out. Get Streamers to play it. Pray that it sparks the DQ

Viral social media

Regular tweeting doesn’t get you too far. Each tweet usually only gets maybe a couple wishlists here or there. But if you can go viral, that gets thousands of people to your page and can start the DQ placement. 

Examples of this are 

A perfect example of this is the game FREERIDE and the developer behind it said “The spike began simultaneously to the wishlist and visit surge on Feb 7th, 2022. The reason was the viral TikTok. During the 2 weeks of the surge, we received 56,460 visits and 15,356 wishlists.”

I wrote about how Freeride had a viral TikTok which earned them over 715k views. 

Actions:

Going viral on social media is tricky. It seems like only games with a really interesting visual style or a very funny hook are able to catch that wave. Horror games usually do pretty well on social media. I always advise trying social media hard for a month or two. If after that time nothing seems to catch, move on. There are other ways to trigger the DQ. 

Running a Kickstarter

Don’t ask me. I don’t know Kickstarter. 

But if you do run a Kickstarter, make sure to have a Steam page live at the same time. 

Here is what one developer said triggered the DQ:

We had a Kickstarter launch for our game (Baladins) on the 5th of May 2022.

Jean Nicolas

I think part of it is the Kickstarter traffic (if you get featured) but also probably a new trailer launch, a bunch of social media, probably the developers contacting their mailing list. The launch of a Kickstarter is a “Beat” which generates traffic. 

Appearing in a festival

For a while I thought DQ featuring only occurred from external traffic such as Youtube, TikTok, or IGN. But many developers from my survey reported a DQ bounce after a festival. 

“Gamescom Steam event with priority placement” was the reason one developer was able to trigger the DQ. 

Another developer said 

“Was part of Steam’s Visual Novel Fest and Love, Ghostie was featured at the very bottom of the page in the Upcoming section. There were four games featured on the first “page” of that section and Love, Ghostie was there the whole festival, which resulted in a lot of wishlists. First day of the VN fest resulted in 2000+ wishlists alone.

The week prior was also fortunate due to being part of the Wholesome Games sale and Tiny Teams, which resulted in ~2000 wishlists total.”

Janbeh Games

Actions you can take:

Apply for festivals! I report festivals in my newsletter and list them all on a spreadsheet maintained by the HTMAG community: howtomarketagame.com/festivals

When the DQ doesn’t happen

The first thing to realize about the DQ is that it is a fickle beast. You cannot expect the DQ. I have seen lots of devs have success doing all the things above and yet, the DQ never showed up for them. 

I have seen devs use the above tactics to induce the DQ one time but then they used the same tactics later, they went viral, or got streamer coverage, but DQ didn’t show up again. It was a one time thing for them. 

WHY?!?

A couple things I suspect. 

FIRST, DQ featuring is based on ranking all games based on tags. So if you are a horror game that has a viral TikTok in October, you might not get DQ featuring because there is another horror game that went even MORE viral than you in October. The DQ only has about a dozen slots, and most are reserved for games that are available for purchase, so there just isn’t room for 2 pre-release horror games.

I helped out the Zero Sievert team (full blog post here) with their big launch and looking through their data they 2 major pre-release DQ spikes. The graph below shows it. Note that the big spike at the end was their launch. 

Having looked at the data for enough big games, it kinda seems like you only get a few shots at the DQ pre release. It is almost like the DQ remembers. My hunch is that Valve is worried you might be shown too much and people get tired of you so they work in some sort of time out period for appearing in the DQ. 

So if you got 2 DQ appearances already, consider yourself lucky and don’t expect any more. 

Can you buy your way into the DQ?

As stated above, it typically takes 1500-4000 wishlists to trigger the DQ. So could just buy a ton of Reddit, Twitter, TikTok, and Facebook ads to trigger it?

I don’t think so, or at least I don’t think it would be worth it. Let me explain…

It depends on the game and the time you buy, and how good your ad-buying-expertise is, but I typically ballpark ads out to $1.50 per wishlist.

That means you need to spend around $2,250 to $6,000 to trigger the DQ. Why not? A couple reasons why I don’t think it’s worth it

  • As stated, getting a lot of wishlists doesn’t always trigger the DQ. So you might be spending that money and see no pickup.
  • I surveyed a lot of developers and nobody said they got paid ads to trigger the DQ. Same deal with clients I have worked for. I have never seen ads trigger the DQ. Not saying they can’t, but I have never seen it.

So you can try. But I think buying ads with the explicit purpose of trigger the DQ is risky.

But if it works and you have evidence of it, email me. chrisz@howtomarketagame.com

Final thoughts

One of the most consistent questions I get is from developers wanting to know how to get out of the rut where they are only getting 1 to 5 wishlist per day. They want to know how they can get the type of visibility experienced by devs who earn 100s of wishlists a day. 

When I look at their data I typically see that they never had a major spike of visibility. 

To truly get promoted by Steam and start earning 50 or more wishlists a day, you need to make a BIG splash. That splash brings in thousands of wishlists, wakes up the DQ (and other Steam widgets), and Steam really starts to show the game off. That is how you get up to 100,000 wishlists in time for launch. 

There are just a few ways to make a “splash”

  • Launch your store page with a bang utilizing a great trailer (see this post on IGN)
  • Have a viral social post (typically on TikTok since Twitter doesn’t work as well anymore, or Reddit like this game)
  • Get a streamer to play your demo (the earlier the better, so get that demo up)
  • Get into a festival (I always share open festivals on my newsletter at howtomarketagame.com/free)
  • Start a kickstarter (you are on your own here, I don’t know it)

You can’t just grind your way to thousands of wishlists with a couple minor tweets every day. It has to be HUGE.