Do the top selling games get all the money?

[James C. Smith]

A recent discussion prompted me to look into exactly what percent of the game sales revenue is generated by the top selling games. Some people have reported that casual games portals the breakdown is like this:

Top 5 Games 35%
Top 10 Games 60%
Top 20 Games 75%

This was according to the DFC Intelligence and CGA Casual Games Study. I have not seen that study and don't know what data is it based on or how long a period of time they are talking about. But I decided to do a much more specific study of the data in Reflexive Arcade.

The fallowing is based on the actual sales data from the last 12 moths of game sales in Reflexive Arcade. In a 12 months period, the top selling 10 game accounted for 29.5% of the gross revenue generated in that period by all games combined. The top 20 games accounted for 42.6% of the revenue. To get to 75% of the gross revenue you have to include the top 86 games.

Reflexive Arcade 12 month period
Top 5 Games 19.6%
Top 10 Games 29.5%
Top 20 Games 42.6%
Top 30 Games 51.0%
Top 40 Games 57.3%
Top 50 Games 62.4%
Top 60 Games 66.6%
Top 70 Games 70.2%
Top 80 Games 73.2%
Top 86 Games 74.9%


If you look at last month (September 2006) rather than the last 12 months it is slightly more top heavy but still not nearly as bad other sources report. In a 1 month period, the top 20 selling games accounted for 54.0% of the gross revenue generated in that period by all games combined. I sampled a few other months and found very similar numbers.


In other words, each month, half the money is generated by the 20 hottest games that month. But over a 12 months period, half the money is generated by the 30 hottest games of the year. This is not nearly as top heavy of other reports that say 75% of the revenue is generated by the top 20 games.

[joe]

good to know

great stats btw

thanks for posting!

[soniCron]

Don't you guys have that great recommendation system? It's possible that is skewing your results because you're delivering less-popular games to people who'd appreciate them. For example, when I'm listening to music on Rhapsody, I rely far more on the recommendations they give me than the top charts. 3 out of 5 recommendations I haven't heard please me, where only 1 out of every 15 top hits do. This means I'm far less likely to "trust" the top charts than a recommendation system, thus confining myself to lower tiered music and bringing up the bottom end.

Of course, I'm the king of speculation, so... Thanks for the stats, James!

[Sharpfish]

Thanks for sharing. Interesting stats!

[Phil Steinmeyer]

Excellent data - thanks James.

[svero]

I guess it depends a little how the portal promotes games. Some sites focus their customers towards hits more than others. On some sites it's hard to see anything other than the top 10. There aren't good features for browsing etc... Reflexive is probably slightly less top heavy than other sites due to the way the reflexive arcade system works and the way your web site is designed.

[tentons]

Reflexive is probably slightly less top heavy than other sites due to the way the reflexive arcade system works and the way your web site is designed.

So, I wonder, does that mean more money for Reflexive than some portal that concentrates more on hits and a possibly less diverse customer base? Are we talking about the long tail here, where there might be more sales from a more broad audience due to more niche sales? Or does it pretty well even out?

[svero]

So, I wonder, does that mean more money for Reflexive than some portal that concentrates more on hits and a possibly less diverse customer base? Are we talking about the long tail here, where there might be more sales from a more broad audience due to more niche sales? Or does it pretty well even out?

That's the million dollar question and one that that portals should be trying to answer. If portals push a game with lower conversion they risk serving downloads of a game that doesnt sell, where a better competitor would have made a sale. On the other hand if they only push a small segment of mainstream products they risk alienating certain customers and not achieving the full sales potential of every product they release. Currently it seems most portals have decided its best to push the few games that sell best. That may in fact be the way they make the most money.. I don't know. They'd have to experiment. Ideally every customer that comes to the site downloads the absolute best game suited to their tastes and therefor that game has the most chance of converting. But how do you funnel people intelligently? Even if I go to a site and I like strategy games or shooters more than match 3 titles, and its easy for me to browse the kinds of games I like, it may still be more profitable on average to show me Zuma over the crop of fairly lacklustre games that makes up most of the shooter and strategy market on the portals today.

[arcadetown]

Here's breakdown of all of 06 and Sep 06 on ArcadeTown.

Top 5 Games 18% / 29%
Top 10 Games 31% / 41%
Top 20 Games 47% / 57%
Top 30 Games 58% / 67%
Top 40 Games 66% / 73%
Top 50 Games 71% / 77%

Perhaps just showing ArcadeTown, Reflexive, and Reflexive affiliate sites give a wider promotion space than other destinations. Looking at your GameSalesCharts World Map Top 10 I'm seeing a decent variety but honestly not a huge amount. It seems users really do want the top games for whatever reasons.

Even so we've had authors comment they liked our production even though their game didn't hit monthly top 10 level so there is life beyond the top few games. Even our #17 spot produced 100+ units last month and for all of 06 we had 101 titles do 100+ units so that's not too shabby even for the lower producing titles.

[arcadetown]

Currently it seems most portals have decided its best to push the few games that sell best. That may in fact be the way they make the most money..

Given they do it and are producing huge #s it could well be. We're going to test the waters with our upcoming new sister site (more on this later).

[svero]

I'm really surprised at the lack of portal experimentation that's been done over the years. If you compare other online sales services like amazon you see a lot more dynamism in terms of new things being tried. There's still so much missing from the marketing equation in terms of online game sales that could even be brought right out of retail game sales and adapted for the download market. It doesn't even require any creativity, yet it's not done. Just off the top of my head, if I'm playing a puzzle or adventure game on Real, why can't I click a button next to the game in the UI and pay a few bucks for a hint guide when I'm stuck? How come I can't ask to see a list of games by the same author? etc.. Lately it seems Bigfish are the only ones pushing any boundaries. Considering how far along the market has come it's really barely moved from the most basic shareware model.

[Bad Sector]

Well, back when Apogee and Epic Megagames were using the shareware model, they were trying lots of things. Then they became big (3D Realms, simply Epic), they went to retail and no new things were tried. Actually, things got removed (f.e. they used to sell a strategy/hint book for every game, they used to give a whole game episode as shareware, they used to not only refer but also promote the game developer, etc).

It could be just that the market had always these people, including a few small guys who were smarter and were driving the business which later became big guys because of their innovations, left the shareware market and only those who weren't "innovative" enough remained. Which actually chopped up the whole model.

I mean... why people don't give a whole game episode anymore, for example? The newest game i remember doing it was Duke Nukem 3D and nobody can even think that this game failed (actually this method proved to be the key of success for Apogee). Or why no hint books? Or why they hide the developer who made the game?

Today publishers (portals in this case) don't only not try new things, but they refuse to do things that others did at the past and proven successful.

[Dyno Kid]

Far too many clones on portals, all match 3 and diner dash.

I contacted a lot of portals and because Dyno Kid was original was rejected by many.

Oberon and Big fish are the only 2 so far that have been brave enough to offer thier customers something original.

Darren.

[Bad Sector]

I contacted a lot of portals and because Dyno Kid was original was rejected by many.

Oberon and Big fish are the only 2 so far that have been brave enough to offer thier customers something original.

It may also have to do with the "frozen shit" colored keys, red outlined green arrows (with huge saturation), badly pixellated tiles and non-uniform look. In other words, with bad graphics. Portals are ruled by non-gamers which look after graphics.

I find your "too many clones, no originality" complaint true and i agree. But while i do, it doesn't mean that those who decide what goes in and what not in portals agree too.

[James C. Smith]

If portals push a game with lower conversion they risk serving downloads of a game that doesnt sell, where a better competitor would have made a sale. On the other hand if they only push a small segment of mainstream products they risk alienating certain customers and not achieving the full sales potential of every product they release.

I believe that aggressively pushing the current top 10 selling or top 10 converting games is a good way to optimize your profits in the very short term. But I also believe that it will hinder growth in the long term. The goal at Reflexive is to always provide the best customer experience. Ideally the customer would download the game he enjoys the most even if it isn’t the game he is most likely to purchase. Certainly the same 10 or 20 games cannot possibly be the best customer experience for all the potential customers.

I do think that most of the big portals are optimized differently then Reflexive and that could have something to do with why they are more top heavy. But mostly, I flat out down believe those numbers. I do not believe that 70% of the sales are from 20 games. Maybe that would be true for one particular portal in one particular week. But many have interpreted those numbers to mean what happens on all the major portals combined over a 1 year period. That is certainly what was implied when this information was presented at Casuality 06. I simply don’t believe that. The World Map shows a more diverse picture than that and is only a snap shot of 1 moment in time. Average over a year I am sure there is more diversity than that. Its is very top heavy. You do see the same games a lot. But I just don't believe the same 20, over a year, would be 75%.

[KNau]

Really? I don't find it that hard to believe. One of the things I've noticed over the past year espeically is that the top 10 list barely moves. The same 5 or 6 titles will occupy the same spots for months at a time.

Look at Mystery Case Files: Prime Suspect; #1 on both Big Fish and Real Arcade and in the "most popular" on Reflexive and has been there for going on 2 (3?) months as far as I can tell. By the time it's off the top 10 it will have occupied a top 5 position for at least 6 months, if not more. Hell, the previous Mystery Case Files game still hasn't left the top 10!

We know the revenue difference between the #1 spot and the #10 spot can be exponential, so when the top 5 is consistently the same 5 titles jockeying for position I can believe that the majority of real income only comes from them. Not even top 10, I'm talking about the top 5 titles being what keeps the portals swimming in money.

[OT] It's driving me nuts. There was an article that used Amazon.com's own numbers to disprove the Long Tail theory but I can't find it. Conveniently none of the longtail desciples seem to even acknowledge it. Basically it boiled down to Amazon making a lot of money off back-catalogue items but the majority still came from the hits. The tail didn't outrank the body, so to speak.[/OT]

[Sharpfish]

It may also have to do with the "frozen shit" colored keys, red outlined green arrows (with huge saturation), badly pixellated tiles and non-uniform look. In other words, with bad graphics. Portals are ruled by non-gamers which look after graphics.

I agree. I also think production values were more an issue than "rampant originality" regardless of what they may have told him to keep him happy, but dynokid does keep hammering that "originality" point home in every single post about this game, so therefore it must be true.

[cyrus_zuo]

I'm really surprised at the lack of portal experimentation that's been done over the years.

I would say, at least for our part, that we are always trying different things. Sometimes they make it past discussion to be added to the list of what we want to do. Sometimes they make it off the list and are added to the website. Sometimes they do well enough to stay on the site. Sometimes they don't.

A couple of things on the Reflexive side, for example, the recommendation system has undergone MANY changes, tweaks and modifications to better match games to people. The ratings on the games have been simplified and allow visitors to quickly rate games, which then is put to use in determining other games that match up. There are a couple of complex formulas being used to match up games to people. It's not perfect, but it has a LOT of time and attention. I think people would be shocked to see how much we've invested in making it better, but we are really committed to trying to match up games to individual people.

Another difference is the way we promote games. The 'favorites' list on Reflexive is based on conversion instead of sales. Sales lists are self-perputating. You show the people the game, they buy it, so it moves higher on the list and it is then harder for a game that is not on the list to displace it. Conversion isn't perfect, but selling more copies won't put you high on the list unless you sold more copies per download than another game.

Coming up with ideas is fun and exciting. Choosing between them and executing them well is also fun and exciting (with a touch of stress). Change has a tendancy to be measured and gradual when it comes to business so that you don't harm the functioning core. I'm really excited about what the future holds .