Category: Updates

Our wishlist estimates have been the best in class since we launched them back in March. Well, they’ve just got a bit better. We have had to tweak our methodology slightly and this has resulted in even smaller variance VS actual numbers in our internal checks.

Key changes:

• Some games weren’t tracking wishlists properly before and they are now
• Some games’ wishlists were overestimated before where we’ve now fixed that bias

This impacts all wishlist numbers we show. We encourage you to update any analysis that relied on the old numbers.

Note that we try and keep any large scale estimation changes to a minimum, but in order to provide you with the best estimates possible, we do occasionally need to make these changes.

We will make sure to give a heads up in the future rather than report any changes retroactively.

We’ve just released our estimates for daily active users (DAU) and weekly active users (WAU) for Steam games.

You’ll be able to see these metrics for any game with at least 100k units sold by going on their individual games page and using the “historical data” section.

Note that these metrics do tend to give a highly correlated view to the much used concurrent users (CCU) metric on Steam. However, DAU and WAU are much more used in corporate environments and serve as a more practical input for most modelling purposes.

Additionally, the DAU metric sets us up for the upcoming PlayStation data where you will be able to start comparing PC & PS DAU metrics and see what the platform mix looks like for multi-platform games.

We have revamped our filter system, making it easier to use and improving its functionality.

Filters are available on many pages, from Steam Analytics to Games Database and can help you narrow down a list of games or companies to look at. We’ve long had an extensive list of filters to choose from, but we’ve now made it even more powerful to use.

Include / Exclude Filters

You can now include as well as exclude characteristics in filters. For example, you might want to see all 4x games, but exclude any turn-based games, leaving only real time 4x games in the list.

Search for Filters

Some of our filters had tens if not hundreds of options to choose from. We have now added a search bar to make it easier to look for the particular item you’re interested in. For example, you can now find all the RPG tags with ease.

We have officially released a long requested feature – Player Overlap. You will now be able to see what other games players play for majority of games on Steam.

Example: NBA 2K23

Using the example of NBA 2K23, we can now see that NBA 2K players are more likely to also play other NBA 2K games (not surprising). But they’re also 20x more likely to play PGA Tour 2K23 (because there was a bundle c. 1 year ago). Interestingly, they are also much more likely to play Madden than your typical Steam player. That is mostly due to US focused playerbases for both games.

What do these metrics mean and what do I do with them?

Firstly, we’re showing overlap metrics for both lifetime sales (LTD) as well as last month’s active users (MAU).

LTD units sold overlap shows any Steam player who owns both games.
MAU overlap shows any Steam player who has played both games in the last 30 days.

Overlap itself shows actual # of people who own both games. Eg 394k people on Steam have both NBA 2K23 and 2K24.
Overlap % shows what % of all NBA 2K23 players own other games. Eg 18.4% of NBA 2K23 players have also bought NBA 2K24.
Overlap Index, perhaps most important of the 3, shows how many times more or less likely an owner of a game is to play other games. For example, an average NBA 2K23 player is 30x more likely to own NBA 2K24 VS your typical Steam player.

Methodology and Caveats

This is a new feature and still in its Beta phase. We’ve tested it extensively, however, there might still be edge cases to iron out. We’re confident that it is directionally right for majority of games, but we will continue to monitor it and tweak where needed. Note that all of these metrics are estimates based on sample sizes and therefore should never be treated as 100% accurate.

Methodology

Overlap % – We look at a large sample size of Steam players to see what games they play. In the NBA 2K23 example. we look at what % of players of NBA 2K23 in our sample play, for example, NBA 2K24. This is used to calculate the overlap %s.

Overlap – We then apply the overlap % to particular game’s (in this case, NBA 2K23’s) lifetime sales.

NOTE that because of that, the absolute overlap # for NBA 2K23 with NBA 2K24 does not match NBA 2K24’s overlap # with NBA 2K23. However, they are in the same ballpark and directionally similar. This is a fluke of this methodology and we will be looking to improve this going forward.

Overlap index – We take the NBA 2K23 overlap % with NBA 2K24 and divide it with Global NBA 2K24 owners %. In this instance, we know that 18.4% of NBA 2K23 owners own 2K24, but only 0.61% of all Steam players own NBA 2K24.

The Overlap Index is therefore 18.4%/0.61%=30.2

Example 2: Helldivers 2

In this example, we can see that these overlaps also work for very new games. The MAU overlap also exposes how popular games tend to overlap with other recently more popular games. For example, Fallout 4 is the 3rd most popular game to be played by Helldivers 2 players in May 2024 as a result of the recent Fallout TV show success. Other recent release, Content Warning also makes it to the top 10.

Who is this for?

This is a feature currently only for our Business Subscribers.

It’s designed to help larger studios and publishers to better understand their own playerbase. It can be used to optimize marketing targeting, to understand cannibalization and relative market shares between competitive games and to see what type of game elements and features your players might enjoy more.

Reach out to us at sales@vginsights.com to learn more!

We have significantly increased the accuracy of our player insights – average and median playtime as well as the country split of units sold.

1. Increased accuracy – Median playtime and country splits for games are now more accurate across all games
2. Quicker insights – Playtime and country split metrics used to take 1+ months since launch to be available. They will now often be available as soon as a day after the launch of a new game

Going forward, we will be able to also start showing how playtime and country splits change over time.

This is also a precursor for the long anticipated Player Overlap feature, launching in the coming weeks.

Games database now includes more metrics for our business subscribers

Business subscribers will now find 4 new groups of columns they can add to the tables – units, revenue, # of promos and average CCU by year.

Users can now easily see and compare how games have performed over the years.

All of the data is available as Excel downloads for the Business subscribers as per usual.

This will significantly improve the ability to do custom research and analysis with our datasets. We’re excited to hear how you’re using the new capabilities to drive business performance.

We’re pushing a major new update to the platform – wishlist estimates.

All Indie and Business subscribers will now be able to see how many outstanding wishlists any game has.

This includes:

Track weekly wishlist change in the Top Charts:

You can now see the biggest wishlist gains in the Top Charts, just like you already do for units sold, CCU and other metrics.

See how many outstanding wishlists any game has

Outstanding wishlists now appear in the individual games pages AND historical trends (but only starting from March 6th 2024 right now).

What about accuracy?

We’ve tested our estimates, and for majority of games, the accuracy tends to be in the same +-15% margin of error that we also aim achieve for our other metrics.

We’ll be publishing public test results soon!

Meanwhile, we are collecting datapoints to continue testing and refining our estimates. If you see large discrepancies, please do reach out!