I've been eyeballing eSellerate's integrated sales (http://www.esellerate.net/eseller_integrated.asp) technology for use with our game. I'm almost done on the implementation. It actually went faster than I expected (system programmer scares the pants off me). Anyway, in the process of reworking on demo period I'm now considering changing up our logic for the demo period. Currently, we have a fairly standard 60-minute time trial. The only check for how much time is left is on the title screen; you can play as long as you want when you're in a game (can play a two-hour game with 1 minute left, etc).

With the in-place purchasing technology, what I'm considering is displaying a demo timer at all times during the trial period. When the timer hits zero, the Buy Now nag screen would immediately appear. At this point, players can either buy the game to continue playing or quit. Has anyone tried limiting the demo period in this way? I've only seen games with similiar setups to our current, where time checks are done at game launch (partly because of using wrappers as protection, but even custom-coded games don't kill in-progress games). I'm worried there could be a backlash of sorts. But, then again, people are playing a trial version of the game; they should expect it to ask them to buy.

Obviously, what I'm after is snaring impulse buyers. If their second game is almost past the point where they hit game over on their first try--and the game stops and prompts them to buy--they're probably more likely to whip out the CC. At least, that's my hope...

I don't tell players how much time or how many games they've got left (simply because I don't know how to explain it in words of one syllable). Players have 1500 seconds or 6 goes - whichever comes later. Then the next time they start a game it just takes them straight to the buy screen. So far it appears to be working excellently.

Cas :)

Crimsonland uses similar approach. I think it works fine.

I have something like that in Charlie II, the trial time is 100 minutes, or actually 6000 seconds of actual play time in the game. Now and then, there is a message displaying how much time remains (first every 15 minutes, later every 5 and then 4, 3, 2, 1). And at 0 the game just stops whatever you're doing.

I wonder, is time limiting the best idea? Everyone seems to do it these days, so it must get results, but I wonder...

If, when you put the hard sell on, you haven't convinced someone within that 60 minute period, that they definitely must buy your game, then your demo is probably going to go straight into the recycle bin; no second thoughts, no second chances. And they can't show your game to their friends... or have several people try the game on one computer. Basically, the demo will not circulate at all once it leaves your server.

I could be wrong (probably am) but it seems that games these days are all too eager to close the sale; perhaps before I've fully made up my mind. As a customer that makes me wary that perhaps there isn't too much to the full game. If a game only gives you one hour of playing time, does that mean the developer is worried that you'd become bored with it if you played it for 2?

It seems that the demos for games are getting less and less generous, and I think it may be because the games themselves are getting thinner. Kind of undermines the whole "try before you buy" thing a little...

Originally posted by Anthony Flack
I wonder, is time limiting the best idea? Everyone seems to do it these days, so it must get results, but I wonder...

If, when you put the hard sell on, you haven't convinced someone within that 60 minute period, that they definitely must buy your game, then your demo is probably going to go straight into the recycle bin; no second thoughts, no second chances. And they can't show your game to their friends... or have several people try the game on one computer. Basically, the demo will not circulate at all once it leaves your server.

I have several demos that I play constantly, and yet, I have yet to purchase the full version. I haven't purchased the whole game because I feel I get all the gameplay I want out of the demos. I feel that shareware demos give you too much time to tell you the truth. I think it should max out at 30 minutes. Or, if you're going to define it by levels, allow me to play 5 levels total. 3 beginner type levels, an intermediate level, and a hard one. I hate when demos just give you the first few levels because it makes it appear as though there isn't much involved in the game.

The whole of my game is available at demo time. What a terrible shame that by the time you get to see the second boss it will probably be your last go :) But I agree wholeheartedly - if nothing stops you from playing the demo after a while you can just keep playing the demo. And you're also right about the "first few levels" syndrome - some people find AF trivially easy (like my brother who reached level 14 on his first ever go, or my wife who got to level 5). You'd want to be sure there's stuff to see in the later levels and be allowed to see it.

Has anybody got concrete figures for "fully-featured time/usage limited demo" versus "infinite plays of feature subset"? Maybe a breakdown by genre? That'd be a nice thing to see.

Cas :)

Good tip on level contain Gmicek!

We'll probably try your suggestion with our INVASION pin.

Thanks,

I've been thinking about this for Werlin. My thinking is similar to a few of the previous posts, that without ANY sort of time/use limitation the customer has nothing pushing them to buy at all (except getting the new features), but alot of people think "Oh there's still time left to buy it, maybe I'll do it next week". But if the demo is telling them "2 Tries Left!", then if they want the game, they will buy it instead of procrastinating and forgetting about it later (IMHO)

I used the example of Doom before and I still think it's an interesting one; because with the shareware version you got a whole 3rd of the game, no time limit, and it was a huge amount of gameplay. It meant that the Doom demo was a worthwhile thing to have in itself and it found its way onto just about every harddrive on the planet. There was certainly enough gameplay there to be happy with... and yet, once you got through that first episode, you were well and truly hooked, and you HAD to have the other two episodes. I think if the demo was made to "self destruct" after 60 minutes or whatever, a lot of people would never have got to play Doom in the first place. I mean, that demo was a must have. It must have been on every school and university computer, everybody's friend's computer, etc. etc. You played the shareware version of Doom, right? But you probably didn't actually go out and get the demo yourself.

Of course, you have to have enough incentive there to make people want to purchase the full version - and this depends on the kind of game it is too of course - but my feeling is that rather than making the full version extra appealing, the trend is now to make the demo version extra crippled to compensate for a thin game. If you feel you get all the gameplay you need out of the demo, perhaps the full version just doesn't have enough goodies to tempt you. Of course it helps if it's the kind of game where you slowly make your way through the content from start to finish (like Doom - I doubt people would be content to just play the very first level over and over again trying to do it faster). If I downloaded a demo that only gave me 30 minutes of trial time, my thought would be "this has got to be a small, shallow game; they're trying to get me to buy quickly before I realise that this is all it is and change my mind". It's also possible that if someone is happy with just the demo, then maybe they don't really like the game all that much and it's fair enough that they don't purchase. I don't really expect to take people's money if they only kinda think the game is sorta okay. But then some games are small and simple by their very nature, I guess a time limit is only natural here.

But it does concern me that demos are being made to "die" so quickly, not only because you might not have time to fully make up your mind, but also because it pretty much limits the exposure you'll get to one person, one time, and that's it.

It would be a difficult thing to test though. Time limited demos must make testing your marketing strategies easier because you either make a sale TODAY, or you miss that sale. So your feedback is pretty quick. But I wonder, it might improve your sales today, or this week, but are you losing sales overall when you count in months or years, and factor in the extra potential exposure you'd get from a desirable demo? But how would you test that...?

I'm tossing up what strategy to go with - my instinct tells me to give a generous (but quite comfortably completable) demo, no time limit, try to get people hooked, and get people to tell their friends, spread the demo as much as possible, and really try to impress upon people the juicy irresistability of the full version, with LOADS and LOADS of cool extra stuff that you just GOTTA have. I'm really counting on the game having sufficient hook to get away with this I guess.

Everyone wanted Doom because it was a revolutionary 3D FPS game. If id has released a two level demo and put a time limit on it, I'm sure it would have sold just as well (perhaps even better).

My experience agrees with Anthony's take. If your game is the kind where there are discrete "chunks" of the game, like levels in Doom or Dweep, providing a limited set of playable chunks works well in the long term. We have people buying who played our limited set for a full year before making the purchase for the larger set.

This was discussed quite a bit a few years ago in the Palm OS development forum. Many agreed that an unlimited, playable subset of a game was the best solution, if your game supported that type of setup. The theory behind why this worked so well was that people kept the software on their Palm devices instead of deleting it. In the extended time of using it they became more endeared to the software. They also shared the software with other potential customers the longer they had it.

Our software has used this formula for years. We had a free version with 3 card games and a pay version with 25 card games. Now we have a free version with 5 free games and a pay version with 50 games. The free version has no time limits, play limits, or crippled features. It is a complete piece of software unto itself that contains an ad for the pay version. The ad is placed at the bottom of the New Game screen. It gets seen every time a new play is started, unobtrusive but noticeable.

Because of this we have, literally, millions of installations of the free version. Most of these people will not convert to customers, but the longer they use our software and not someone else's, they become more attached to it and share it with others. Also, the company name gets the benefit of being considered a "contributor" to the market because we give away what many consider to be a robust and feature-rich game at no cost. It is robust and feature-rich because it is a subset of a for-pay game that I spent lots of time polishing.

However, if your game is not one that breaks up into chunks well, you obviously cannot use this. The formula that was agreed upon for that type of game was to limit the game time over time, rather than a general time pool. This means that after the initial 10 days or whatever, the game is limited to, for instance, 15 minutes of play in any 24-hour period. Numbers would have to be adjusted for the particular game. The idea was that the demo would stay on the user's machine if they were not interested in purchasing at that time, and that keeping a demo on a user's machine for as long as possible was desirable for long-term success.

-Tom

Anthony, I do share your concerns about the demo "dying" quickly, but I recently decided to try a time limit for Charlie II anyway, because I have a feeling that just giving one third of the game away for free is too much.

It's hard to tell exactly how much effect the time limit has, since I also changed some other things and new releases usually seem to increase sales for a while. But to give you an idea, I released the new version at the end of November and in December I had over 60% more sales than I was having in the months before. So it looks like it really works, but of course I don't know what the long term results will be.

For my next game, I'm not sure what to do. I sure do plan to have a smaller percentage of the game in the shareware version, probably 1/7th or less. I've also been thinking of having a limited time per week or per month (like 30 minutes per week or 10 trials a month or something like that).

We use a combination of usage and limitation.
It is limited to 5 plays before expiring and limited features (no custom name high scores, no labs, can only play on 15 day length).

It works well and I definately agree that most shareware games offer way to much. When you combine both limitation and amount of play time, it really encourages the customer to register - as not only can they continue to play what the demo offers them as much as they want, but they feel they get a bonus when they get even more features unlocked.

Alex

Many agreed that an unlimited, playable subset of a game was the best solution, if your game supported that type of setup. The theory behind why this worked so well was that people kept the software on their Palm devices instead of deleting it. In the extended time of using it they became more endeared to the software. They also shared the software with other potential customers the longer they had it.

For one of our Palm games, we allow users to play for 3 "in-game days," then display a "buy now!" screen; however, if the user re-enters the game, he's allowed to play for another "in-game hour" before receiving the notice/ejection again. A few people who have posted mini-reviews on PalmGear and Handango have said things along the lines of, "you can keep playing for an hour at a time, but I eventually had to buy the whole thing, because I couldn't stop playing."

Aside that, I wonder if you can play the numbers game with demo length. They say that a $9.95 sale price is better than $10.00. It strikes me that...

9 Trial Levels -- 50 in the Full Version!

...is better than...

10 Trial Levels -- 50 in the Full Version!

...because it seems like you're getting more in the former case.

My current ideas for my demo includes:
Showing as many aspects of my game as possible with actual gameplay.
Limiting the length of the game not by time but by gameplay (the game ends at a certain point in the story)
Showing very clearly how big the demo is compared to the full version. ("You have just finished 10% of the full game. BUYBUYBUYBUY ;))
I want the player to have as much fun as possible playing the demo. I have no doubt that having a 60 minute time limit hanging over your shoulder removes some of that fun.

I think an important part of making the demo short/long enough for the user is to make sure the player can actually finish the demo.

Mike: I have seen a novice play Charlie the duck, and she could barely finish the first out of the three levels. After a few hours practice she could get past the second level. When I, a veteran platform player, played, I came to the third level after a few tries and met a boss that I know a novice player couldn't defeat in weeks. A player will hardly buy the full game before even completeing the demo. Or to put it another way, any person that is not good enough to complete the demo is very likely not to buy it. Hiding a too difficult game behind a timelimit might work to some extent, but a well balanced game should let anyone progress at a decent speed anyway thus removing the need for a timelimited demo.

Mike: I have seen a novice play Charlie the duck, and she could barely finish the first out of the three levels. After a few hours practice she could get past the second level. When I, a veteran platform player, played, I came to the third level after a few tries and met a boss that I know a novice player couldn't defeat in weeks. A player will hardly buy the full game before even completeing the demo. Or to put it another way, any person that is not good enough to complete the demo is very likely not to buy it. Hiding a too difficult game behind a timelimit might work to some extent, but a well balanced game should let anyone progress at a decent speed anyway thus removing the need for a timelimited demo. That is Charlie I, which is a lot harder than Charlie II actually. I''ve received so many e-mails, letters and even telephone calls about how to defeat the first boss (the big fish), so I decided to make Charlie II easier. Just getting to the exits shouldn't be a problem at all in Charlie II, but now you have to collect a certain amount of items before you may exit. If you set the difficulty to easy, you only have to collect 55% and you can easily get through the shareware version in less than 10 minutes and for an experienced player it's possible to get 100% at all six levels in less than 30 minutes. Of course, it will take longer for someone who is new to this type of games, but 100 minutes should be enough to decide if you like the game or not IMO.

Which is fine if it's just you, but is it long enough for your partner to decide, your kids to decide, your friends to decide?

And as Tom says, even if someone is happy to just keep playing your demo, that still shouldn't be seen as being without value to the developer. They know your game, they know your company. They like you. Maybe a year from now they will be playing the demo and think, "I wonder if this developer has any new games out now?". And playing your demo for free is still better than purchasing a competitor's product.

Having a time limit that resets is an interesting idea; sounds more healthy than having a demo that dies for good. If someone likes your game they're not going to settle for an hour a week, or 15 minutes a day, or whatever (especially if it cuts them off mid-game)

Simonh - Doom was a revolutionary title, true. Which means that it had a high hook factor... once you saw it, you wanted it. Thus they could afford to have a generous demo, knowing that they would soon have a huge fanbase of hopelessly addicted players. Having a fat demo ensured many more people were exposed to it and thus more people were hooked. With a time-limited demo, who knows, maybe it would never have reached critical mass... as it was, the whole WORLD knew about Doom - and I don't think this can be attributed to its technical merits alone.

Maybe it took people 15 hours of playtime to finish the Doom demo... but with Doom people happily played for 15 hours and still wanted more. That's the difference I think - you could not confidently say that about the majority of recent shareware games.

It makes me wonder when people say that shareware games give too much away... maybe it's true in Charlie's case, certainly if you can't complete the demo you won't want to buy the full game - but if that's the case the game is too hard for you and you wouldn't get to play it properly anyway. But the recent trend definitely seems to be to give away very, very little. Which is why I say, if very, very little is still too much, then there can't be much substance to the whole. And if less truly is more, perhaps it's an indication that developers are losing faith in the whole shareware concept - perhaps your instict is telling you to abandon shareware altogether and just have people pay to download, period.

I should have an interview up on indiegamer soon that has a question to someone "famous" about this very thing. IE the doom trilogy approach vs. the 1hr very limited demo. Should be interesting to see what he says.

I'm not so sure more stringent restrictions are as much a lack of faith in the shareware model as they are recognition of the power of impulse buying. The idea of dealing out trial minutes on a daily or weekly basis intrigues me (you could go further with it too--what about rewarding someone who uses a "tell a friend" feature with more trial minutes?). I wish we had the 1st-party traffic to experiment with various methods. Currently, our own site traffic is too small to get any sort of useful data on changes.

I am wary that longer trials or renewed trial periods are a double-edged sword, though. While they may be a great reason to keep the game installed, they could also be used as justification against a sale. Why buy now, when you could save the money and play for a little tomorrow? I'm hoping to ensnare impulse buyers with our new demo version--it lets you keep playing as soon as the purchase is complete. Play is interrupted when the demo timer hits zero, at which point players can launch in the in-application purchase process or quit. I deployed the new demo online, if anyone’s curious how it looks in action: http://www.flashbangstudios.biz/buzzwords/installBuzzwords.exe

Oh, and on the subject of trial variations: I've heard that a 60-minute trial is over 20% more effective in conversion than a 5-use execution trial. I believe I heard this at last year's GDC, but I need to check my notes on who claimed that. We're simply using a 60-minute trial because it seems to be what everyone else is doing; I'm assuming there's good reason it's a standard of sorts.

I find some of Matthew's points interesting.
Originally posted by Matthew
I'm not so sure more stringent restrictions are as much a lack of faith in the shareware model as they are recognition of the power of impulse buying.

This may very well be true. In the Palm OS discussion about unlimited play for a subset of game chunks, it was an assumption that any particular game that was successful with the approach was a game that did not necessarily hook the player in a short time. Games that can hook in a few minutes may indeed be better off with an impulse-buy model.

Originally posted by Matthew
I've heard that a 60-minute trial is over 20% more effective in conversion than a 5-use execution trial.

This is the kind of data I would like to find out more about, assuming the data is scientifically sound. I wish I was in a position to attempt to generate some. Wasn't Steve performing a download/no-download test on Dexterity at some time? It looks to me like he may still be testing, as Blastorama does not have a free demo download for me.

As another anecdotal example, I really enjoyed the Crimsonland demo. I played through the demo quest segment in one go, and actually dreamt about playing it that night. I told some people at work about it the next day. This was one I just knew I was going to buy. I would play through the demo quest segment again once or twice to try some things I had thought up, then try some of the other modes to see what they were like, then I would be hooked and buy it to see the other 4 quests.

I was completely surprised that night when the demo timed out. This was new to me, since I had not encountered timed demos much in the Palm OS world, and certainly not for a game this good with such obviously discrete play chunks. To me, it begged to be an unlimited demo with restricted levels. The other 4 quests were locked out, for goodness sakes!

Now I've uninstalled the demo since I can't run the demo any more on my machine. I will probably forget how fun it was. I did not uninstall out of anger, I did it because the demo no longer served any purpose on my machine. I did not buy simply because the checking account I use to ration purchases like this needed another few days before I put more money into it. In my admittedly singular case, the timeout prevented a sale.

I present this not as an argument that Crimsonland shouldn't have used the timeout, but as an argument that without any quantifiable data about impulse buying versus long-term trials, one method can't be known to work better than another. In the end you really do need good data on the various methods. This points out, to me at least, that Steve's article on testing your methods is very important. I wish I had implemented some testing methods for our Palm software, because a few tweaks to the demo might have brought in more profits.

If anyone knows of any available research data on demo restriction methods, I for one would certainly love to hear about it. :)

-Tom

When marketing my game BLOX I received many comments that it lacked a timed demo. It seems the market place has migrated hard to this.

Having played Popcap's new game Zuma, I may well agree with it. The player gets "Zuma Deluxe" for free with a 60 minute demo. The demo was great and had no limit on levels so I progressed pretty far into the game only to come up against timer wall. Thinking hard about dishing out that $25. Nuts!

The reason I think giving one third of a game like Charlie II is too much is that you're actually giving about 80-90% of the gameplay. What people buy are only the other levels in which this gameplay will take place (of course including new graphics and music). New kinds of enemies and various new game elements will make up for the remaining 10-20% of gameplay itself.

I think this makes the incentives too low. With the time limit, you eventually get the choice between 100% of the gameplay or none at all. I might experiment a little with the amount of time, maybe 180 or 240 minutes is better for a larger game like this?

I agree with both Brian and Mike on this. This was also the consensus reached in the Palm games discussion.

In games where, as Mike points out, 80% or more of the gameplay is experienced in the demo, a timeout is best. Zuma is a good example.

In games where the gameplay is different in the unplayable part of the demo, such as locked levels in Dweep or Doom, unlimited demo play is best.

Successful games seem to bear this opinion out.

Does anyone know, even anecdotally, of the success of locking out large portions of the gameplay and using a timeout? Crimsonland uses this, and my understanding is that Crimsonland has gotten a tremendous number of downloads but the conversion rate is really low. Is this true?

-Tom

As I, in fact, complained about Zuma (although I didn't say the name at the time)...

My hour demo timed out while paused because I stopped to watch TV. So I got far less than an hour.

My hour demo timed out on my first play without me having ever lost a single life. (Admittedly, I didn't get as far as some people, because of the above problem.) There was never a challenge, and I certainly didn't get to see what the other play modes were.

So, my impression becomes - This game is easy and pointless, and these people are trying to steal my money by not giving me the chance to fairly try out their game without paying for it.

Chances of sale out of me? 0%. :)

Limited plays rather than strict limited time would have given me a chance to fairly evaluate the product.

Sure, it may indeed work better at hooking people who are not very bright (smug snobby response, but I hope you know what I mean... The Impulse Buy is in itself not very bright) and a lot of times in online business you *want* to target the person who is not very bright, they're much easier to make money off of. Porn sites, frex, are desperate to get traffic from AOLers, who are seen as much better pigeons than the average net user.

Actually that happened to me playing Zuma too. I never got to finish my first game. I paused it for a bit cause I had to go feed the baby. Didn't realise that that "counted" in my hour.

Other play modes? Were there? Who knew?

My plan with my own game is to give away just three levels of easy mode, two levels of medium mode and one level of hard mode, plus the tutorial mode, and not impose any time limit. Total levels available for free - 6/60, or 10%.

I think this gives people enough of a taster of what the game is like, without giving a lot away. Just a short sample.

I don't want to impose a time limit as I think time limited demos are far less likely to stay on a hard drive then unlimited play ones - and even if someone doesn't want to buy my game, I would still be happy with them keeping it on their hard drive and playing it from time to time.

Yeah that's pretty much how I feel too. Although in my case I'm planning to give away 12/60 main game levels (the first of five chapters) and probably 3/15 bonus games (subject to change possibly since I haven't actually made most of the levels yet :) )

But in my case it's necessary to give away quite a few levels in order to fully impress upon people just how many different kinds of levels there are. They will all be quite different from each other you see...

Was it Platypus that limited you to 3 plays? I seem to recall the following train of thought:

Try 1. "Damnit, I accidentally exited out instead of playing."
Try 2. "Hot damn, I can't believe I'm playing a claymation shooter! Oops, gotta go somewhere."
Try 3. "Hmm. I'd better save the last try for when I have more time to play it."

...which really prevented me from spending "just a few more minutes" with an otherwise cool product. From the sound of it, you're taking another approach with your upcoming game.

It was Platypus' publisher that added the SoftWrap software to it. Earlier demos didn't have that restriction.

And, in fact, the first version of the original demo was quite a lot longer; the publisher insisted I cut it back. And then next thing I know, the softwrap version comes out. All nowt to do with me.

Originally posted by Anthony Flack
And, in fact, the first version of the original demo was quite a lot longer; the publisher insisted I cut it back. And then next thing I know, the softwrap version comes out. All nowt to do with me.

I think a good demo of platypus would be the entire first world including the first boss. My initial impression of the game was very poor based on idigicon's demo. It was actually bert the barbarian that sold platypus to me.

That's about the amount of content I like to give away, because it's about the amount you need to show people how much content is in the whole thing. The 1 level Platypus demo is such a small fraction of the game, just the very very first few things.

Still, a lot of people still can't finish the first world :D

Originally posted by Anthony Flack

Still, a lot of people still can't finish the first world :D

Yeah the game needs some work to sell well online. Skill levels and better out of game menus especially. It could do much better than it's probably doing now I'm sure. It's too bad because there's a lot of really cool stuff in the game that most people never see I'd imagine.