Which games are actually being played in internet cafes?

You want a ranked list. Ten titles you can install this week, confident that those are the ones putting people in seats. I understand the impulse and I am going to spoil it in the first paragraph, because handing you that list would be the least useful thing I could do. It does not exist. Nobody has failed to compile it out of laziness. Cafe play is local in a way that makes a global ranking meaningless.

A title that carries a room in Turkey barely registers in Brazil. Regional publisher clients that never appear on a Western storefront can be most of a room's play in one market and unavailable in the next. Regional pricing, patch schedules, whether a game even tolerates a shared machine and a shared account, what the local scene competes in: all of that decides your list and none of it appears in a worldwide chart.

And the charts people quote at you are measuring the wrong people anyway.

Why published charts cannot answer this

A public concurrent-player chart counts everyone equally. Someone in their bedroom on a machine they already own counts the same as someone paying you by the minute. Those are not the same behaviour and pretending they are is how owners end up with a library nobody touches.

The person at home plays what they own. They installed it once and it never leaves the drive. There is no clock running, no reason to choose. Your customer left the house, paid for time and is now deciding what to spend it on. They pick things that justify the trip. Whatever that is in your city, a home-player chart cannot see it.

So stop hunting for the list. Three lists actually exist. Each answers a different question and only one of them is genuinely about cafes.

List one: your own Applications tab

Open Statistics and go to Applications. Apps ranked by cumulative usage time across every client PC and every session, usage grouped by category such as games, browsers, communication and productivity, plus a trend view showing how that shifts across the date range you pick.

This is the only list on this page that is definitely true. It is your room, your customers, your machines, counted in minutes people paid for. No survey, no proxy.

It is also the narrowest thing you will ever read. It can only report on software you installed. A title that would have been the best performer in your cafe cannot appear on a list of things you never put on the machines. That is survivorship. It is why the Applications tab tells you about your library rather than about your market.

Read it alongside the other tabs rather than on its own. Sessions gives you session count, average duration, revenue per session and peak hours. Computer Usage gives you per-PC utilization, uptime, revenue per machine and your busiest PCs. Consoles does the same for console types. An app sitting at the top of your usage table on machines that are never busy is a very different fact from one sitting at the top on your peak-hour machines.

Here is the part most owners skip. Do not read the top of that list. You already know what is at the top, you can see it from the counter. Read the bottom. Anything sitting near zero for a month is shelf space, patch bandwidth and possibly a paid account you are still renewing. The Client Menu has a visibility toggle. You can hide a title without deleting it and see whether a single person asks where it went. Nobody asking is your answer. That is the cheapest audit in the building and almost nobody runs it.

List two: the Games catalog, with the country selector

This is where HandyCafe stops being a passive recorder and starts pointing outward.

The server has a Games catalog page, part of the Cloud subscription. It browses in rails: Most played right now, Trending games and New releases. There is a country selector with a Global option. The data behind it comes from Steam charts and Steam storefront metadata, with artwork from SteamGridDB.

Set the country to yours. The rails change. That is the whole point of the feature and it is the difference between reading a worldwide chart and reading something shaped like your market.

What it tells you honestly: what is live right now where you are, what is climbing and what just launched. It catches things before a customer walks up and asks whether you have it, which is a much nicer position to be in than the reverse.

What it cannot tell you: it is Steam. Steam is home players. It is closer to your market than a global chart because of the country filter. The population underneath has not changed. Plenty of what runs in cafes is not on Steam at all. Launcher-exclusive titles, regional publisher clients, browser games. Those are invisible to anything derived from Steam and their absence from a rail says nothing whatsoever about how popular they are in your city. If your room runs on something outside Steam, this list will never mention it.

The catalog earns its keep as an installation tool regardless. Adding a game from it pulls in cover art, poster, genres, rating, release year and a description in the client's language, then the title appears in the client menu on every connected PC immediately. Even when you already know exactly what you want, that beats doing it by hand. If you are building a library from scratch rather than auditing one, the reasoning behind what to install in an internet cafe is a better starting point than any chart.

List three: top items in cohort and its honest limit

Cloud Benchmark compares your cafe anonymously against a cohort of cafes in your country. Your figures stay anonymous, you only ever see aggregated peer numbers and other cafes never see yours. Most of what it shows is money and occupancy: percentile bands for revenue and utilization against the cohort median, plus an hourly occupancy profile across the day.

Sitting inside all of that is a list called Top items in cohort: the top games, apps and products across cafes in your region.

That is the one. It is cafe data from cafes. Paid seat time, not bedroom time, from rooms with roughly your customers and roughly your problems. Of the three lists on this page, it is the only one that answers the question you actually asked.

Now the limit. I am putting it here rather than in a footnote because it decides whether the feature is real for you. The cohort is built once at least twenty cafes in your region contribute anonymous data. Below that, the tab tells you the cohort is still forming and shows suggestions based on your own numbers instead. In a country with a dense cafe scene you will have it. In a thinly covered market you may be waiting. While you wait you are back to list one. That is the honest position. There is more on what the cloud side is doing with this data in esports cafe data in the cloud.

Where owners lose the afternoon

Install things because a chart said so.

Every title you add costs disk, patch bandwidth, sometimes a paid account and always a bit of staff attention when it breaks. A chart cannot tell you whether one person in your room wants it. Adding six games off a trending rail is a satisfying afternoon that produces a library nobody asked for and a slightly slower update cycle forever.

The deeper trap is that library work feels like business work. If your seats are empty on a Tuesday afternoon, your game list is almost certainly not the reason. Rearranging it is motion that looks like progress. The actual problem is usually that not enough people know the room exists or have a reason to come at that hour, which is a getting customers problem and not a catalog one.

Churning your library every fortnight has its own cost too. Regulars want the thing they played last week to be on the machine they sat at last week. A cafe that constantly rotates trains people to expect nothing to be there.

What I would do

Start with Applications over a stretch long enough to include both a school holiday and an ordinary week. Cross it against Computer Usage and Sessions so you know whether your top titles are top at your busiest hours or just top overall.

Then open the Games catalog with your country selected and look only for things that are climbing which you do not have. Not the top rows. The gaps.

Then check Top items in cohort if it has formed for you. Where it disagrees with your own list, treat that as a question worth asking rather than an instruction worth following. Sometimes the answer is that the cohort has a title you should test. Sometimes the answer is that your crowd is different and that is fine.

Cloud AI Suggestions is worth a mention here, not because it picks games but because it does not pretend to. It reads your own data over a period you choose and comes back in plain language: revenue up or down against the previous period, your best day, your weakest weekday, your peak hours, your net profit margin and tips such as a quiet weekday window worth promoting. It will never tell you which game to install. It will tell you which hours are worth filling. That is the better question.

Then walk the floor and ask people. No dashboard has ever beaten the counter for this.

And if the underlying worry is whether the seats will fill at all rather than what is on them, that is a bigger question and it has its own honest answer.