★ TouchArcade needs your help. Click here to support us on Patreon.

Cricket Words Mid-Mortem

07-30-2012, 12:41 PM
#1
Joined: Nov 2011
Posts: 94
Cricket Words Mid-Mortem

Ultimately this post is about getting suggestions from other devs. Let me give you the background. There are questions throughout and I've included user numbers and profit data.

We launched Cricket Words on November 22, 2011. Words with Friends was at a height of popularity, but we had various issues with it which I won't go into here. Our idea was not to make a Words with Friends clone. Far from it. We just wanted to make an awesome word game that we'd actually enjoy playing.

The app is here if you want to check it out while reading: http://bit.ly/cricketwords

Game

Our game would be a combo of word-building games like Boggle with scoring from the dart game cricket. The game revolves around making words of particular word lengths. For example, make a 4-letter word and you get a mark for 4s. Make three 4-letter words and you have cleared 4s. Make another 4-letter word and, as long as your opponent hasn't cleared them, you score 4 points on them. The winner is the first to clear all marks with the lower score. To further tie it to cricket we included ways to get doubles and triples based on playing a red letter on a red square.

In order to avoid the cheating that is rampant in Words with Friends (one of our main gripes with that game) we added a timer. Once you start your turn your letters are revealed and you have 45 seconds to make up to 3 words max.

Does this all make sense? One red flag might be how hard this game is to explain. We included instructions in the app but that may not be enough, people might not read them, or the game still might not be clear after reading them. We were trying to do something different but I'm wondering if we were a little too different. This is something I'd love feedback on.

A few more notes: the game allows you play against 2, 3, or 4 players. We launched as a universal app for iPhone and iPad. We included a number of Game Center achievements and leader boards right from the start to track total wins and points as well as various accomplishments.

We iterated extensively throughout the dev process, playing with our friends all the while to test and see how people reacted to each change. Things like:

- We originally had 4 red letters available to make doubles. This made the game far too easy.
- We originally only had one difficultly (what ended up being Hard). We could see certain people were struggling and decided we had to launch with an Easy and Medium option.
- UI went through various stages, as I'm sure is normal.

Business Plan

We went through various ideas as to how we would position the app to make money. We decided against just charging for the app. Charging for a multiplayer game from a company that had never released a game before seemed like a bad idea since it would be too high a barrier of entry to expect your friends to also pay to try an app that you like.

Ads and in-app purchases were the obvious options.

We added advertising. At the time we implemented AdWhirl and set it up to use iAd, AdMob, and our own house ads (directing users to our other apps) for when those couldn't be filled.

We initially had a handful of in-app purchases. Ad-free, of course, which lives on to this day. We also had in-app purchases for being able to play against more than 1 opponent and the ability to play more than 1 game concurrently. Ultimately we felt these last two were just hindering people from playing and would lead people to abandon the game quickly. We scrapped them both before launch and allowed anyone to play 2/3/4 player games and to start as many games as Game Center would allow.

For interstitial ads, we did not want to be as annoying as Words with Friends, displaying an interstitial after every turn. Instead we would display an interstitial once every three turns. On the iPhone the interstitial was essentially an advertisement to purchase the ad-free upgrade. On the iPad we attempted to use the iAd interstitials if available, otherwise we'd display the ad-free upgrade option.

Pre-Launch

The game would be multiplayer and use GameKit's new (at the time) multiplayer mechanics and matchmaking. This meant that we were launching as 5.0-only very shortly after the release of 5.0. We didn't think too much about this since using GameKit's built-in functionality was really the only way we were going to get this done in a reasonable timeframe. Potential downsides were excluding people who hadn't upgraded to 5.0 and people who couldn't upgrade to 5.0 (because their device didn't support it). A potential upside was hoping that Apple might want to feature an app that was iOS 5-exclusive (they didn't). We blogged about this, posted some data, and then our numbers from when we switched to allowing 4.3 and up. As you can see, didn't really make a difference.

We had a number of friends and acquaintances playing Cricket Words from the moment we had something playable. Prior to release to the App Store we posted on a few forums (including here) with screenshots and promo codes (the game had been approved but we hadn't released it yet). We honestly didn't and still don't have much of a Twitter following, though we tried to ramp up how much we tweeted and posted to our blog ahead of launch.

We sent close to 50 requests to review sites to try to get some kind of attention. We included promo codes so they could download the game ahead of launch. This resulted in exactly zero reviews.

What could we have done better pre-launch?

When trying to post I'm being told my post is too long. I'll break it here and post more about the Launch and beyond.
07-30-2012, 12:42 PM
#2
Joined: Nov 2011
Posts: 94
Launch

We updated our site. Sent our tweets. Blogged. Posted in forums. Sent an email to our list. We advertised Cricket Words within our other apps.

Launch day came and needless to say we were disappointed. We didn't exactly know what to expect, and we certainly did NOT expect to be in the top charts or making $100s per day. But it was free and it was a game and we figured we'd get at least 100-200 the first day. What we got was something else entirely. If my memory serves we had 40 new players that day (added to the 25-30 we had prior to release). We were very disappointed. In retrospect we probably didn't do nearly enough to promote the game ahead of time. I'd love to hear how people think we could have better promoted Cricket Words.

We always kept track of users based on how many players we had in Game Center as opposed to downloads, since that could be misleading. Now, about 8 months after release we have 2,800 players (of course not all active). On last check we had about 100-200 active players in a given month.

What could we have done better at launch? We didn't buy advertising. We kept saying, "once the next version is released" or "once feature X is in there". Never did it.

All in all we figure we spent about 200 hours getting the initial version of Cricket Words together.

Updates

In the beginning stages we were issuing updates constantly. Occasionally they were bug fixes, but more often than not they were gameplay improvements, UI improvements, new achievements or completely new features. We would always post on these forums with updates and for awhile we had some active discussion with players. That's since died out.

Around the time of our launch, Apple was moving updated apps to the top of the New Releases list. This has since been done away with. To be honest, we were fond of it because it gave us a good incentive to keep making meaningful updates to the game. Every time we'd update we get a good boost in our numbers. And that would make us want to keep going with more updates. It was a good cycle. Obviously, I can see how some were probably abusing this App Store rule, but we never felt like that was the case for us. Eventually Apple changed the policy on this and apps now do not go to the top of list when being updated. Ultimately, this has given us less incentive to keep updating because we don't know how to get a bigger audience. That's where I'm hoping you can provide some guidance.

One significant update was 1.2, when we decided to finally reverse course on requiring iOS 5.0 to see if that would make a difference. We changed the minimum version to 4.3. Since 5.0 was still required in order to do multiplayer we came up with a 1 player variation. This quickly became the thing people did most often in the game and it also was where we could show the most ads. Although this didn't really increase our number of downloads : number of players ratio like we thought it would, it ended up being one of those "we should have launched with this / we're idiots" moments.

Although we don't get the update boost any more we certainly haven't abandoned Cricket Words, and that's why I prefer to call this a midmortem rather than a postmortem. Recently we released with 1.4.4 with Retina graphics for iPad and various bug fixes. Just today we've submitted version 1.5 to Apple that includes a new Extreme difficulty mode (this is an in-app purchase or unlock for free with 25 wins). Of course, we don't see any boost now when we release an update.

Profits (or lack thereof)

In the spirit of transparency, here is some information on how the game grew and shrunk and related profits on a month by month basis. Believe me, I know how sad these are.

11/2011
  • New Players: 600
  • iAd: 30.62
  • AdMob: 6.44
  • LinkShare: 8.00
  • In-Apps: 16.48
  • Total: 61.54

12/2011
  • New Players: 1000
  • iAd: 50.12
  • AdMob: 17.48
  • LinkShare: 7.00
  • In-Apps: 16.80
  • Total: 91.40

1/2012
  • New Players: 565
  • iAd: 9.72
  • AdMob: 4.75
  • LinkShare: 2.00
  • In-Apps: 6.30
  • Total: 22.77

2/2012
  • New Players: 135
  • iAd: 10.92
  • AdMob: 3.19
  • LinkShare: 4.00
  • In-Apps: 6.30
  • Total: 24.41

3/2012
  • New Players: 200
  • iAd: 17.56
  • AdMob: 2.62
  • LinkShare: 2.00
  • In-Apps: 4.20
  • Total: 26.38

4/2012
  • New Players: 100
  • iAd: 11.87
  • AdMob: 5.42
  • LinkShare: 1.00
  • In-Apps: 6.30
  • Total: 24.59

5/2012
  • New Players: 100
  • iAd: 11.25
  • AdMob: 1.97
  • LinkShare: 1.65
  • In-Apps: 1.85
  • Total: 16.72

6/2012
  • New Players: 100
  • iAd: 11.14
  • AdMob: 0.93
  • LinkShare: 0.50
  • In-Apps: 2.61
  • Total: 15.18

Total
  • New Players: 2,800
  • iAd: 153.20
  • AdMob: 42.80
  • LinkShare: 26.15
  • In-Apps: 60.84
  • Total: 282.99

A couple of notes here. 11/2011 basically means 11/22 - 12/21, etc. for each following month. We had our ads targeting iAd if at all possible. LinkShare is included because we would always include our LinkShare affiliate ID when promoting the app and our house ads pointed to LinkShare links. We recently changed our ad-free in-app from being $2.99 to $1.99. We have about a 1.1% in-app conversion rate, which seems low compared to what I'm seeing people post.

We were excited that we had an increase from 11/2011 to 12/2011, modest as it was. We were hoping that trend would at least continue. But as you can see there was an immediate drop-off after the holiday period (when iAd eCPMs were really good and we were still attracting a decent number of new users every day). Things have continued along a relatively steady line after that, slowly declining.

Things We've Learned

It seems to us that the business of the app (ads vs. purchase vs. in-apps) is probably fine. That's not really the problem. The problem is people aren't using the game and/or aren't sticking around. Of course we think the game is great, but that's a given. It's our baby. I'd be highly indebted if you'd take a few minutes to try the game and provide feedback. What's confusing? What do you like vs. what don't you like? Is this the kind of game you play? How do you feel about the UI? Where are the holes in our implementation? This kind of feedback is highly valuable to us.

We learned we should have launched with our 1 player mode. By the time we included it it was too late to make as big an impact as it could have. Someone trying a new multiplayer game probably wants to know that they enjoy the game before "recommending" it to a friend by inviting them to a game.

We thought our game was launch-ready but in retrospect we're embarrassed by version 1.0. We should have waited to launch until the game was truly awesome. As one example, take a look at how our UI evolved from 1.0 to the current version:

http://blog.thinktapwork.com/post/27...s-ui-evolution

Next Steps

So what's next? We've submitted 1.5 and we still have lots more ideas for how to extend the game. Some of these involve new in-app purchases. But without an audience it won't make much difference.

Perhaps relaunching with a publisher would make a difference. Honestly, being new to game development when we started we didn't even know this was option. Does anyone have recommendations of indie game publishers that you like or have used yourself?

Conclusion

Think Tap Work is a mobile app development company. While in time we'd love to be able to survive on revenue from just our own apps, at the moment we spend most of our time doing client work. That having been said, Cricket Words was, for us, still:

- a fun project to do
- an education in GameKit and game development in general
- a great portfolio app

We would probably do it again be I'm sure we'd do a ton differently. Thanks for reading and any feedback you can give.

Jeremy Provost
Think Tap Work

07-30-2012, 11:16 PM
#3
Joined: Jul 2012
Posts: 15
Hi Jeremy -- I'm a former game journalist and current game designer, so I thought I'd give a little feedback on both levels. As you suggest in your 2nd post, there is a problem in your complexity level. Here are my moment by moment impressions:
  • First screen: iOS blue box with three options. Immediate bad impression; a really well-made game wouldn't start this way.
  • I chose "Instructions" and was treated to four screens with diagrams. Read about 3-4 words then skipped on.
  • Checking options on the front screen. This is actually nice. I'm clear on what I want: single player...
  • OK, now I'm playing. What do I do again?
  • Successfully entered a word. I win! I got a dash. Huh?
  • Backgrounded.

I like to play through games with most of my attention directed elsewhere -- in this case I'm watching the Olympics. If I know what's going on and have fun, the game is successfully designed. Games that can't pass this test are almost never successful, and if they are it's only with a hardcore audience.

You've probably heard it a hundred times before, but on mobile you do not have your player's full attention, and they definitely will not read and absorb an instruction manual. Not with several thousand free apps out there that require less effort. This counts about as much for a blogger as for an ordinary player -- either one has a lot demanding their attention.

I did go back and play a bit more, and the good news is that the game actually seems fun (which is probably why you at least had some minimal attention and downloads -- one of my past games failed much worse than this!). Best advice is to get in some basic analytics and start testing out tutorials that simultaneously require player participation and tell them exactly what to do, preferably lasting under a minute. Look at what the big companies do to teach players, because they spend massive amounts of time optimizing. Don't feel embarrassed to copy what works until you get your first success; at the end of the day, your actual game is still unique.
08-02-2012, 08:24 AM
#4
Joined: Nov 2011
Posts: 94
Quote:
Originally Posted by cmo View Post
I'm a former game journalist and current game designer, so I thought I'd give a little feedback on both levels.
First off I want to thank you for taking the time. The kind of feedback you've provided is 100% exactly what we were looking for.

Quote:
As you suggest in your 2nd post, there is a problem in your complexity level.
Kind of what I was afraid of. Unfortunately, at this point it seems like a tall-task to make it a simpler game (unless I'm missing something). But the other suggestions you give are certainly a step in the right direction. If the game can't be simpler at least try to make it simpler for the user to understand.

Quote:
First screen: iOS blue box with three options. Immediate bad impression; a really well-made game wouldn't start this way.
Nice observation. Never really thought of it this way, but you're right. To be honest, this is a holdover from version 1.0. In our first release there really wasn't much of anything on screen when you first opened the app and didn't have any existing multi-player games. We've since redesigned that aspect of the UI but never took this dialog out. Great suggestion.

Quote:
I chose "Instructions" and was treated to four screens with diagrams. Read about 3-4 words then skipped on.
1 point for honesty. Agreed, tough to get people to read.

Quote:
  • OK, now I'm playing. What do I do again?
  • Successfully entered a word. I win! I got a dash. Huh?
  • Backgrounded.
I have to admit, I laughed a little when I read that. I'm sure that's the same process 75% of people go through, if not more.

Quote:
Best advice is to get in some basic analytics and start testing out tutorials that simultaneously require player participation and tell them exactly what to do, preferably lasting under a minute. Look at what the big companies do to teach players, because they spend massive amounts of time optimizing. Don't feel embarrassed to copy what works until you get your first success; at the end of the day, your actual game is still unique.
Very good advice. We'll look into this. Thanks again for trying it out and sharing your comments.
08-02-2012, 09:32 PM
#5
Joined: Jul 2012
Posts: 15
Yeah, actually ignore me on the complexity level of the game itself. That's probably fine, or even net positive, for people who like word games or puzzle games anyway. IMO the devil is in how you introduce it. Good luck!
08-08-2012, 03:22 PM
#6
Joined: Nov 2011
Posts: 94
Cricket Words 1.5 with Extreme mode just got approved and released to the store. Next up we're going to focus on these things that you mention. We'll be working to improve the first impression and hopefully make it easier to understand and start playing.
08-08-2012, 05:48 PM
#7
Joined: Jul 2012
Posts: 15
Congrats! Actually we just had a new version of our testing app go live too, and as it happens it's one with tutorial improvements. We've only had a couple hundred users through so far but are seeing a 3x improvement in reaching a key point in the funnel (from 22% > 70%).

So.. it is possible to make massive improvements, especially when there is something users view as broken in the initial experience. In our case it was requiring Facebook Connect to access some key features (dumb move) but it can also be as simple as introducing gameplay correctly.
08-08-2012, 05:59 PM
#8
Joined: Jul 2012
Posts: 20
cmo gave you some really good pointers, so I'll only add a few different ones,

the screenshots in iTunes, look incredibly dull, like a blackboard with chalk, reminds me of school. (Not fun).

If you look at other successful apps, they are colorful, have particle FX and stars flying everywhere etc. While one thinks that a word game shouldn't need it, from my perspective that is exactly what is needed. So that even costumers who are not into these kind of games, (like me) will at least download and try it, just because it looks fun and maybe even enjoy it.

Another thing would be advertising, I know that some developers don't believe in it, but how will people find your app if they don't know it exists.

And lastly, if you are afraid of something, usually it will happen or don't go well, trust your instinct till you remove that fear factor. That's just how life is according to my experience.

Hope this helped a little bit
08-09-2012, 06:59 AM
#9
Joined: Nov 2011
Posts: 94
Quote:
Originally Posted by cmo View Post
Congrats! Actually we just had a new version of our testing app go live too, and as it happens it's one with tutorial improvements. We've only had a couple hundred users through so far but are seeing a 3x improvement in reaching a key point in the funnel (from 22% > 70%).
What's the app so I can check it out?
08-09-2012, 02:05 PM
#10
Joined: Jul 2012
Posts: 15
We're not live in the US Doing the painstakingly-slow-live-iteration thing.