Steam makes discovering new things easier, since "average" sales of "average" games fall by 70%

[ad_1]

Separating the wheat from the tares on Steam is not always easy. Valve's open door policy means that many small developers are lucky not to do otherwise, but it also means that many promising games are lost at random. Unfortunately, according to CEO Mike Noble Robots, the problem of Steam overcrowding has a direct financial effect on "little guy" publishers.

Rose attempted to evaluate the "average game" of Steam sold in July 2019 compared to the same period last year. In this case, the average excludes big triple-A games, games with less than 10 user reviews and the top 5% and lower of the makers. Rose did not have access to direct sales figures, relying on Steam group numbers, revision numbers, prices, algorithms and her own Steam sales experience. So yes, these numbers are only estimates, but they are quite well educated. In short, according to Rose, the average Steam game was sold at around 5,000 copies in July 2018, but only 1,500 in July 2019, a decrease of 70% from the previous year.

Related, connected, related Valve did not mean that Metro Exodus Epic exclusivity comments were a "lightning rod"

I've analyzed the Steam sales data from last month and found that:

• The average game sells for around 1,500 units and brings in about $ 16,000 in the first year.
• This is down 47% from the previous year.

The full report can be found here: https: //t.co/vITIPDUVZl pic.twitter.com/ICvHu7b51b

– Mike Rose (@RaveofRavendale) September 10, 2019

Rose has not responded directly to Rose's estimates, but it is clear that they know that overcrowding is a problem on Steam because the last update of their store only concerns the possibility of discovery. In the future, Steam's recommendation feeds will be less focused on popular big games, but more on content specifically tailored to the interests of that player. Based on beta testing, recommending more discrete and more diverse games encourages more people to click on recommendations …

We want to make sure that we show our customers a diverse set of games while keeping them relevant. But would they commit to these recommendations? Would they click? Would they like these games? Would they buy them? To answer these questions, we've changed the way we display referral code-based customer games in store locations, integrating them with our bug fixes, and sending them to 5% of customers to test them. during the last years. weeks.

In these changes, "Recommended for You" has become less focused on popular games and has shown more relevant games for individual customers. It turned out that clients in the experimental group were more likely to click on the games presented in the recommendations section, at a rate nearly 15% higher than the control group. Increased customization means that the number of games in this section is even larger and customer impressions are better distributed among them.

To get an idea of the magnitude of the titles visited, we measured the number of games that members of the experimental group visited via the "Recommended for you" section compared to a sample of clients who did not have participated in the experiment for a few days. The results were very promising: we saw a 75% increase in the number of unique games visited and 48% of the average number of visits per game.

Hopefully, these changes will help Steam's "middle" game a bit, but the big problem with the platform, that there are too many things, still remains. I do not think Valve will be able to solve this problem with some algorithmic modifications.

What do you think? What should Valve do to make things easier to discover? When was the last time you tried your luck in a Steam game under the radar?