Portals and how to fix them

[svero]

Ok. So here's the problem a portal faces. They have 50 games. They have 10-12 very good promotion spots. In order to make the best of their revenue they have to promote the games that sell the best of the 50 they have. So they work out some formula based on downloads, conversion, and overall sales and try to pick the best 10 to give their limited spots to.

That all makes perfect sense. However... It's pretty clear that if you gave the 11th-14th games a spot in the top 10 they'd do better than they would on page 2. So you've potentially maximized your overall sales by promoting the best games (assuming you did indeed pick the right games to promote), but you've also definitely not made the most of the games you aren't promoting as heavily. Sometimes someone is coming to your site who wants the 11th game but none of the other 10 and doesnt see it and you lose a sale. That's bad for #11 and it's bad for you.

So the question becomes.. how do you maximize the potential of each individual game without hurting your overall sales?

I think some ideas that answer this are tracking customer preferences and displaying different games, and offering more search and sorting features, but really I'm interested to see if anyone else can come up with some clever ideas to solve this problem.

[ManuelFLara]

I've not even seen the RealArcade software running so I don't know what data is displayed and how. But when a user logs in he/she sees the 10 best selling games, even if he/she has already bought some of them ? I mean, why not simply do not list the games the user has already bought? So if he/she has bought zuma, ricochet and platypus, supposing they were at top 10, the 11th-13th games will be at positions 8th-10th.

The key question is: Why would you want to promote games to a user that has already bought them?

[DaveMyers]

How about not setting your sights on a lousy top 10 list and instead expand it some. When I go to a retail store to buy a game I don't see a top 10 list. And even if I did, I still see about 80-100 other games side-by-side with Mr. Top 10 pretty much equally considered. So, why is it that it's okay to accept 10-12 "very good promotion spots"?

Personally, I think you have to question the assumption that top 10 is the best way to attack things first. I believe that a portal like Real is too-shortsighted if they restrict their marketing in such a way. But, hey, what do I know? ;)

[oNyx]

>even if he/she has already bought some of them?

Or even if he/she has already downloaded the demo?

Well, there might be people with a broken download and they forgot about it... or maybe some people who would turn into customers just by a reminder, but there shouldn't be too many of them (like one in a million).

You could also have something like recently viewed/downloaded-list (with maybe 5 titles), which would make getting back to a specific title easier (especially if you forgot the name). I mean... hey... it's stuff like this cookies were invented for ;)

[arcadetown]

This was one of the major considerations in our recent site revamp. We now present more selections on our home page, made our categories more clear, plus some other extras like most played, top rated, additional sorting options, etc. Seems to have helped but sometimes wonder what would happen if went the opposite direction and concentrated on a select few titles like some others do.

[James C. Smith]

That's bad for #11 and it's bad for you.
The font page of most portals has the top 10 sellers, a what’s new list, and some featured game. But most importantly (to your point), they also have many other things to click on. You can click a category to see all the card games. Or click one of the top 10 games to see other games like it. There are probably 50 different games you can see with only 1 click on the front page. I don’t mean you see a list of 50 games. I mean you only see 10 games at a time, but which 10 depends on that one click. It is really not as bad as you make it sounds. The # 17 selling games does still get seen. Part of the reason the #1 selling game sells so much is because of the self promoting nature of the top 10 list. But another big part is that the #1 selling game is a really good game.

I believe that if you made a portal with the font page always picking a game at random you would still see a very skewed sales chart. In other words, give every game in the catalog an equal chance it being shown on the front page. The front page would pick 10 games completely at random and pick a different 10 for each visitor. I think you will still end up with the top 10 games making up almost 50% of the sales. This is just my biased opinion, but I really think some of the games are that much better than the average game. I don’t mean it would be the same 10 games all the time. Each month two or three great games are released that hold the markets interest for about 3 months. You may find that with a completely random front page, some of the great games would continue to sell much longer. But the mediocre games wouldn’t sell much better.

Just look at the difference is sales from the #1 game to the #10 game. They both get nearly the same promotion. They are both on the same page. Yet the #1 game outsells the #10 game by 5 to 1 or 10 to 1 depending on the portal. The #1 game really is that much more attractive to the customers.

[Og!]

svero, I couldn't agree with the sentiment more, but why call it a portal problem? If you only have 5 games, but for someone who comes to your site looking for something different, you're missing out just as much with the portal who has 50. Their depth is an advantage, the question is if they are able to use it (more than to pick the 10 best sellers from that is)

With respect to James C's thing about some games just being that much "better" and the 1-3's selling better than the 7-10's, users are trained to click on the highest link. Every web designer knows how much of a difference putting something first does. I don't run a game sales site, so it's just a guess, but I bet those "much better" top 3's get a heck of a lot more downloads than the 7-10's, so the interpretation of total sales proving their worth might be missing out on some important factors there.

Also, it's impossible to prove, so maybe useless of me to say, but what about audience? I would think that If you promote what sells you attract others that want to buy it. It's a cycle.

[arcadetown]

Good points James. Our new look has spread the wealth and flattened the curve somewhat to the marginal expense of top sellers. Top games sell because they're good games and more exposure means more sales, the ok games do a little more ok with more exposure, while the bottom games won't sell more no matter how much they're exposed.

[svero]

I don’t mean you see a list of 50 games. I mean you only see 10 games at a time, but which 10 depends on that one click. It is really not as bad as you make it sound.

I agree with everything you've said. Clearly if you pick random games the games that don't sell and don't normally get into the top 10 aren't suddenly going to start selling better. That isn't the point. And obviously there are a few ways that users can click and see other games like the new list or categories. Reflexive arcade is probably the site that does it best. All the same, I don't think that changes the basic fact that portals don't get all they can out of every game they have to promote. So yes I'm focusing in on the problem a little and ignoring some of the other details, but that's mostly just to talk about the one problem without discussing every single thing a portal might want to consider, like the best way to promote new games etc...

[svero]

svero, I couldn't agree with the sentiment more, but why call it a portal problem?

Well I'm sure that some of the issues are generally applicable to smaller sites, but the bigger game portals are all still fairly unsophisticated. As has been pointed out, a player who's already bought zuma is still pitched zuma. On a system like realarcade that's bordering on incompetance. The site's been around long enough that those sorts of basic things should have been worked out you'd think.

[svero]

How about not setting your sights on a lousy top 10 list and instead expand it some. When I go to a retail store to buy a game I don't see a top 10 list. And even if I did, I still see about 80-100 other games side-by-side with Mr. Top 10 pretty much equally considered. So, why is it that it's okay to accept 10-12 "very good promotion spots"?

I think it's a practical consideration. If you use up all your screen space on the front and most visited page showing 80 games, you're not really doing a great job promoting them all. There's only so much room on the screen. Think of it as a small store with virtual shelf space and only enough room for 10 boxes facing forward and a little room for 20 boxes shelved on their sides.

[Stu]

I have recently been studying this whole world of game portals. I had never paid it much attention before. I started fiddling with RA just one month ago. I have since played 54 demos. (I just counted them)

Yep, I played the top ten first. Then I played the games with the best seller icon. Then I played anything that appealed to my personal tastes regardless of its' placement within the site.

I did not buy anything for a few days. Now, 30 days later I have bought a grand total of two games and guess what? They are both in the top ten list. The point is that I think James has a point.

If I am not mistaken, all new titles have a shot at the front page in the newest games list. It is directly below the top ten list. I check it every time I log on and I'll guess that most folks do to. Of the ten games currently on the RA newest games list, four of them are on the RA top ten list. It seems to me that where they go from there or whether they got there in the first place is up to the game itself.

Stu

[svero]

I just want to make it clear. I'm not suggesting top 10 titles sell because they're displayed in the top 10. I understand that the top 10 games sell because they're good games. However, that doesn't negate the fact that the 11th game is still probably pretty good, and while it will sell, it's probably not selling as well as it could given the way the current systems work.

[Og!]

The way I think of the issue of the sales divide between just a few spots on the portals & what people choose to promote is related to the whole casual games vs. console vs. traditional PC markets themselves...

I know a lot of girls (grown up now...) that used to play duck hunt and super mario in the NES era - even some that played colecovision stuff... But they stopped games completely by the SNES era. Some of them played some GBA, but nothing compared to the number that play bejeweled and stuff like that now. And ten years ago, how many of us would guess that 40ish women would be playing PC games today? the minority I'm sure

My point being that most of the game markets in the world are pretty limited in terms of audience. So why can't we have a downloadable market that isn't? PC's have a very wide installed base over quite a varied demographic. From my perspective, any top ten list is bad if it isn't right for the individual customer.

Relating this to indie passion, I would expect people here are making stuff that they love and want to share it with as many people that will love it too as possible. The business needs for money seem to make people push the thing most people like - as opposed to pushing the thing a person will like most

(that's kind of why I think it's not just a portal problem - the issue is finding the most people in the audience for every game, a site with a small library may have an customer based well matched to what they sell, but it will be small)

[Diodor Bitan]

Support for A/B tests and a slow releases to allow games to evolve their conversion rate to it's potential may help.

[cliffski]

its true that new games DO get their shot at the top spot, Starlines and Planetary Defense did, and got sidelined when they didnt sell.
However they DID sell hugely better in those days than when they are moved off the list. I can pretty much tell where starlines is listed without looking, juyst by noting monthly sales.
I think that games should be randomly given the odd day in the top 10 list, otherwise regular visitors are always shown the same predictable list (and its usually the same list on every portal). I've pretty much given up visiting portals, precisely because the top 10 list seems the same everywhere.

[luggage]

There was a discussion not long ago about how your game sells well when you're in the top 10 list then it drops away very very quickly as soon as you're out of it. This is exactly what we found on all the portals our game went on.

The cabinets in the UK that we put games on also has a similar situation. It has a main menu that has one big huge button and a few smaller ones, any game on that button area does well. Other games are in categories one touch of the screen away but don't do as well. There's still room for overlap, a poor game on the big button will still do poor obviously.

How about doing an A/B test, decide what the top 10 games are and display them for one set of customers. The other swap out the 6-10 positions and replace them with 11-15.

[Nauris]

I still have to see selling strategy topped by that of Amazon. Their suggested products make sense to me. They must have tagged each product with various genre/setting flags, which means the more i buy/download, the more precise those suggestion picks would be.

There`s nothing more satisfying than finding previously unknown great product which fits your interests pattern.

[princec]

How about simply creating a second site run by the same business, but deliberately not including the 10 top games from the first site, and instead featuring games 11-20 as 1-10? Effectively run two completely different portal brands.

They do have top tens in retail stores. Even Sainsbury's has a top 10. I don't think they base it on sales though - instead I think it depends on how much money they've been paid to put a title in it. The retail world is curiously topsy turvy. Perhaps the portals will follow suit?

Cas :)

[svero]

For a company to run a 2nd site probably wouldn't be very reasonable. It's hard enough building the traffic for one site. The only exceptions I might make to this are language and platform. You might want a 2nd site for mac games or for german games etc... And potentially even for genres. But genres could be handled easily with category links such as those reflexive and other sites already have.

[luggage]

The other thing to consider is that people will think the games in the top 10 are better and than those that aren't.

[James C. Smith]

I understand that the top 10 games sell because they're good games. However, that doesn't negate the fact that the 11th game is still probably pretty good, and while it will sell, it's probably not selling as well as it could given the way the current systems work.

Ah. I see your point. That is a tough one. If you acknowledge that your game truly is the 11th best game, how can you maximize the sales of that game. This reminds me of several business and marketing books I have read. They say is hard to make money being #11 (or even #4). The market leader takes most of the customers and the next two runners up take almost all the rest. To succeed, you need to find something specific that you can be #1 at. You may not have the most desirable “casual game”, but you may have the best “action card game” (or some other specific sub category.) And you are right, this is a problem of how best present these games on a diverse portal. Customers with focused interests need to be able to easily find games that meet those interests. I think this relates to what I was talking about a few weeks ago when I said game portals need to be more like Amazon.com and get better at selling every type of game rather than focusing on the top 10 ultra casual games.

[Andy]

Cross links (used at Amazon as well btw) could help very much here.
But in common what to say - James is right with his books - be the best at least in something specific - otherwise "no way!" :)

How to deliver that message that your are the best in the area? - this is another story really. And any regular portal with their mass market can't help here to the specific original developer/publisher.

[cliffski]

this is very true. I'm aiming for democracy to be the number 1 political simulation game, hopefully also to be one of the top complex turn base sim games.
I'm not even pretending to compete as a general purpose strategy or tycoon style title, because I know its crowded. I was even thinking of trumpeting the complexity of the game as 'the most complex simulation game you can buy'. That probably turns off 90% of casual gamers, but I'm hoping it puts me top of a small niche.
Interesting thread.

[svero]

I think it's definitely true that portals have to be more like Amazon. And probably these sort of questions are what led amazon to do what it did. Although nowdays Amazon is kind of bloated and not everything is presented very well. But proper cross promotion and focusing in on individual customer likes/dislikes definitely goes a long way to solving this problem. Maybe it's the best one can hope for.

Still I wonder about rotating selections and other strategies. Maybe some simple systems with cookies that highlight different titles when a user visits but while maintaining the top 10. The problem with top 10 games is that on average a portal is right to promote those. Just generally as averages go there's more chance they'll make a sale if someone downloads one of the best sellers than another randomly highlit game. But maybe there's something to be said for encouraging more tries. Encouraging more downloads. If I go to realarcade and see the same top 10 games every time then I'm less likely to try a game the 2nd or 3rd time through that I havent already downloaded for whatever reason. Clearly replacing that title might give a site a chance at a sale it wouldn't have had otherwise.

[arcadetown]

Never really been a fan of top 10 lists as they self perpetuate themselves causing static content. Our new look emphasises new games and spreads the wealth. Let the users decide, what does well gets further promotion.

Doing good for our authors, particuarly those with solid niche products that weren't able to give proper exposure to before. Total sales up nicely and saw this change in sales behavior.

top 10 : 73 -> 59%
11 - 20 : 19 -> 22%
21 - 30 : 5 -> 11%
all others : 2 -> 8% (edit)

Not saying this is best for all sites. We get a very spread out representation of users so niche games can probably do better here than elsewhere. Would like to see more shooting, adventure/rpg, and strategy games if you have some good ones.

[KNau]

Unfortunately the major portals won't change until they start losing money and the current projection is that the business will continue to boom until 2007. Personally, I think that things will drop off sooner than that.

The portals probably look at Amazon as a business model and maybe that's part of the problem. Correct me if I'm wrong but hasn't Amazon yet to achieve a profit? Same goes for Real Networks, last I saw the games division was the only one that was making a profit and the rest of the company was in sad shape. You'd think the tech-bubble mentality would have passed but all you have to do is look at Google's stock price to see that the insanity isn't over yet.

[Nonz]

Offer free membership and allow users to customize how the site looks to them.

Obviously everyone has their own tastes in games so someone that prefers FPS for instance would see the top ten in that category (as well as a general top 10 of course).

Allow them to change colours and layouts, maybe even have particular game themes, anything to encourage them to make the site look just the way they want it too.

Also include more than just "Buy my game", things like industry news and events, articles, strategy guides and such.

Basically make it a proper portal, one that people will use as their start/home page.

Nonz.

[James C. Smith]

Amazon may not be (very) profitable, and they may have lost their focus lately, but they do have some very nice tools for finding thing that appeal to individual customers. I love to do all my shopping at Amazon because they help me find things I didn’t know I wanted. Assuming the develops all do good jobs at making a bunch of games that are best at one particular thing, the game portals are going to need to way to help users find those game efficiently similar to the way Amazon does..

Re: RealArcade’s profitability, are you sure they are profitable? It is a little hard for me to image how they could not be, but I also would assume they would brag about it if they were. The Real Networks financial stamens always break out how much revenue was generated by the games division, but they never say how profitable it was. (unless I missed it).

[svero]

Well they shelled out quite a bit recently buying gamehouse so that may have been a factor in profitability.

[Coyote]

I remembered an article about "The Long Tail" of sales in the new points-of-sale that are less limited by shelf and warehouse space in Wired.

I had a couple of thoughts on this - not very deep at all, and some of it may be a reiteration of some things that have been discussed here a few times on the subject. I posted them my blog (http://www.rampantgames.com/blog/2005/02/long-tail-in-independent-games.html).

From the Wired article, it's my feeling that it doesn't make much of a difference to the "near misses." If you are #21 in popularity, well, you are screwed no matter what. A really good system (such as Amazon's) might help introduce you to customers that wouldn't have found you otherwise - but it's not gonna change the fact that your sales are going to be disproportionately smaller than the sales of a game just 10 ranks above you. It's not fair, but it's the way it goes.

What it does show is that there is some money to be made with products that historically haven't made the cut-off because they are too unprofitable. It means that potentially - with a large enough offering, deep price cuts, and good searching / recommendation tools - the top 10% of the games no longer represent 90% of the profits, but only, say, 40%-50%.

Now, it might be WAY too much work to sell downloadable games to double your profits. But potentially it might be viable. And that could be a big deal for some of the more niche-y games that wouldn't see the light of day anywhere but their own website.

[Mike Wiering]

I'm not sure how this would affect sales, but I think it would be a great way to list relevant games to the user (though it does involve a bit of spying): let all the games you install log how often you play them and for how long and return this information every time you visit the portal site (perhaps by letting the games modify cookies or something).

I think that the number of times people play a particular game and the total time they spend playing it would be a far better criterium than the number of downloads (since that already depends on the exposure, and on how interesting the little thumbnail looks and how the title/description sounds) or sales (many games seem great for a while and then become boring).

Whenever a user enters the site, it could use all this information to generate a top 10 list (top 100, whatever) of games based on which games other people like to play most, who also play some of the same games this user does (sort of like Amazon).

This way, games that might look a little unpolished but are actually real fun to play get noticed by many poeple. Users would usually get a lot of value for their money and would like to come back to that portal and buy more games. OTOH, they might just keep playing the games they bought instead of buying new ones.

[Alan_3DAGames]

I think one very important issue is what to do about keep selling the same old games, and its a difficult one to solve???.

On the one hand, Indie developers want (and need) to build up a customer base for their games over time ... especially as they cannot afford much or any advertising money etc...

On the other hand, retail games (and many other retail products) don't stay at the top of sales forever. New products come along and older ones naturally don't stay on the selves as they are physical products in boxes etc... so limited shelf space etc...

This situation means that over time more and more games will join the lists of available downloadable games and so each game will have less and less market share. It seems to me to be a big flaw in the current Indie Business Models that over time the market is getting flooded with ever increasing numbers of older games and no way to clear out the older poor quality stuff. So developers (especially the one's who are not so established) will earn less money as a result.... and the situation is currently getting worse.

We can't simply stop selling older games, as some of them may actually be very good games. On the other hand, we can't keep selling 1000's of games as no one will make much money and the potential customers will get sick of diging through sites looking for something that is good when they have dozens of poor or cheaply made games or poor clones of clones of games etc... to look through.

The top 10 lists seem to be a way to partly (and poorly) deal with this issue, but even then a good new game will not sell as well as a older game in a top 10 list, so the new game will be at a disadvantage to even have a way to make it into the top 10 lists of some sites.... so some sites ends up selling the same top 10 games ... and customers want new games they haven't played before, otherwise they will loose interest in the sites.

We need ways to encorage the new games to do well, (then everyone can start earning better money) .... like having the lists of games on sites being ordered by initial release date. (And only initial release dates, not re-releases otherwise that may get abused with small re-releases etc...)

But we also need sites to filter out older good games and keep them in (their own) lists .... maybe also have on the sites a list of shareware classic games as well as a list of games by initial release dates etc.. That way the good old games and new games appear at the top of lists.