Why a paid app is a bad idea

Had this conversation with my wife AGAIN, and she still does not get it. So I actually did the math, and it is worse than my gut instinct told me, so I thought I would share.

I have an app. It has around 30k downloads. Around 2k users, some of which log in and play every day (some play all day!).

The app talks to a cloud server. I used Google App Engine. This costs around $2/day. Advertising income is around $30/day. Good deal right ?

So lets do the math on a ‘paid up’ version:

  1. Since the app could continue from now till forever, I need to cover costs on the interest only (lets assume no inflation!)
  2. Working out how much money I need to cover $2/day as interest only, with a generous interest rate of 0.5% gives: 365 * 2 /0.005 = $146k
  3. Assuming I sell my app for $3.00 and Amazon takes their 30% cut, gives me $2 per sale. However, after state and federal tax, I end up with around $1.
  4. Sooo… I need to make 146 000 sales of my app at $3 each JUST TO BREAK EVEN (excluding inflation).
  5. $438k sales for one person is MAJOR SUCCESS story that you read about in the news… highly unlikely to happen!

My wife says “but any money is better than no money” (since advertisers pay on a net 90 day policy, I have not been paid yet)… umm… no honey, you are dead wrong on this.

To some degree it depends what your ongoing costs are. If you’ve got a game which is entirely client-side (except for perhaps Google Analytics), you’ve got practically no ongoing cost. So once somebody installs the game, it costs you nothing.

In that kind of situation, a paid app would make sense, assuming two things. One - the average lifetime advertising value of a user (total revenue from all the ads this user would click, ever) is less than the amount you’d sell the paid app for (say $2.99); and Two - you are prepared to sacrifice the cross-promotional value of that app.

The main problem I see with paid apps (apart from server upkeep costs) is the lack of cross-promotion ability. There’s incredible value to be gained from promoting your best revenue-generating apps across your other free apps. This is hard to quantify, but can reap huge returns in the long term.


Remember, your revenue calculation goes like this (being loose with the math to just show easy numbers):

App price = $3.
Amazon cut = 30%
Your income = $2
Your income after tax = $1 (assuming 50% tax rate… obviously this varies by country)

Anyway, the point is: DO NOT DO YOUR CALCULATIONS based on projected sale price !

And when it comes to advertising income, you have the following options:

  1. Reinvest it in the app with advertising to further increase your userbase
  2. Invest it in advertising another app (cash for startup mode?)
  3. Get paid and pay taxes

On the subject of taxes/etc. I wonder how many small developers (not me!) created a legal entity in another country (erhm… Vanatu or another tax haven?) so they can get paid without paying local taxes ?

Another part of the equation is how long will the app keep the users entertained for. From your numbers it would take about 4.5 months for 2000 users to generate the same revenue if they all paid for it.

2000 * 2$ = 4000$
4000$/28$ per day = 143 days.

Whats hard to figure out is how many users would pay for it if it was paid only. I imagine it would reduce the number of users as well as the number of days to break even significantly.

Your numbers are very good and its seems like the non-paid model is the best for your app.

An ad-only version also solve the problem of piracy. Although I feel there should be a paid version as a service for those who really like your app and don’t want ads. There are some developers (such as those in countries who can’t sell paid) who only do ad-apps and do very well with them.

I still disagree. I think you would still get your $30 a day from all the people who don’t want to buy an ad free version. Plus you would get extra revenue from the people who buy the ad free version. The people who buy an ad-free version aren’t clicking on the ads and contributing to that $30/day anyway.

Don’t forget, that the download numbers for a free app are much higher than for a paid app. Far more people will try out your app and maybe like it.

As soon, as you have ongoing costs you need to have also ongoing income - thing is, if 10k users paid for your app they want to use it forever - so you have to keep it alive and “forever” pay your ongoing costs - which will not work, you need a steady income.

Personally, I think, that an ad supported free app is currently they primary way to go for some very important reasons:

  1. More people will download your app because it’s free.
  2. You can use EVERY store to publish.
  3. You have no problem with piracy.

A point to mention is for sure, that some users will not like ads and they want to pay for an adfree version - I am still searching for a way to make this possible without harming the points above…

Maybe I’ll dismiss ads in a higher level … hey, that would mean I got more space for user controls! xD

A legal entity in another country would be too expensive for a small developer :wink: You need agents there and even Panama needs bilancing since this year so you have to pay someone to do it there for you.

For the games I’m busy working on I’m looking at using Tapjoy and/or Zooz to remove ads. Zooz will work for all markets (although I think inapp billing is more convenient and trusted for most users at this stage) and Tapjoy works well too although I think it’s dependant on Google Play store so wont necessarily work for all marketplaces.

Surely it comes down to whether or not over the course of a lifetime, you would make more money from a user clicking on ads or a user purchasing the app. My best ad network is currently paying something like $0.15/click. This means that if a user will click on ads more than 14 times in their lifetime then they are more profitable using the ad supported version (assuming $2 for the paid app), with a few caveats. Paid apps get the money up front meaning you can invest it and earn more from it. Ad supported apps, as alluded to above, have cross-promotional potential.

Also, those tending towards purchasing the app probably don’t like ads as much and are probably less likely to click on them in the first place. Hence, have both versions. Best of both worlds. My app is a weather app, and its paid version revenue far out paces the ad revenue. Totally different paradigm to game apps however!

I don’t believe, that a user will click 14 times on an ad. But that calculation is wrong.

Look, a free app got tons of more users than a paid app - you need to consider this when calculating.

So if you look on free vs paid downloads, you will find, that the free version get’s downloaded up to 10x more, than the paid version. So 1 user buying the paid version equals around 10 users playing the free version - in you calculation ending up with 1-2 ad clicks per free-version-user.

Very true. It’s impossible to directly compare free vs. paid downloads, because the difference itself affects the customer psychologically - they’ll approach free & paid apps completely differently. We can make some general calculations, but in real life it’s something that has to be looked at on a case-by-case basis.

This… and then realize that some markets are USED to paying.

Soo… maybe do a paid only app on Nook (since no adverts are allowed) and do advert supported on other platforms.

Then build in an additional monetization scheme…

The calculation isn’t wrong and my own app has many more free users then paid ones. Its based on the assumption that you have both a paid and free version of your app and that most users come in to your app via the free version. You then incentivise them to pay for the paid version (no ads, more features, etc.) if you believe you make more money off of them in a paid version. Some people will never pay or upgrade and that’s fine. You can continue to make money off of those through ads. However, at least for my app, the more people I can shift from the free to the paid version the more income I make from that user over their lifetime usage. Therefore, paid apps aren’t a bad idea. They’re a great idea (at least for me), but the key is bringing in users via the free version and then shifting them to the paid one to maximize the amount of money I can make from them.

another thing to mention would be the type of the app - i.e. time of actual usage. An ad supported game needs to be addictive, and give a user a long time of fun.

Every app which has a defined (short) time of usage will better suit the pay-model to be profitable.

For some games and apps this can be quite a challenge - consider apps which run in the background - there is no great oportunity to monetize them with ads

consider apps which run in the background - there is no great oportunity to monetize them with ads

This is right but i guess push notifications ads can be used to overcome this . I know of a developer who develops only wallpapers and makes a fortune out of it and the users have accepted push ads since he makes quality wallpapers and two the users do understand he cannot integrate a banner ad or so.

Your initial calculation is wrong for so many reasons.
First off you will increase the price in the future to compensate for inflation. So there is no need to put any inflation corrections in your calculation.
Secondly you assume players will play your game forever, that’s not how this happens in real life. Almost all players will stop playing the game within a year.
Then you take in account that players will continue to click on your ads, that’s also wrong. New players are more likely to click on ads, the rest will ignore them over time.
And lastly not every free player will pay for the game.

But all in all, if I take your numbers, $1 after tax and $2 running costs. In short you will need 2 buyers every day. That’s 365 * 2 sold apps per year to break even. Its as simple as that.
And yes in very very very long terms that’s will be a couple of thousands, but I remember you would also need much much more free players in the same period that will install your game, because no player will continue to use your game for years and years and also keep clicking the ads.

Just to illustrate that the above math is also wrong (this problem is hard!).

My current costs are $2/day and proportional to the number of users. So the costs scale with the number of users. You can have fixed income from a user and then use all the income to cover today’s cost… but that user continues to cost you in the future! So this is not a great idea.

Btw, yesterday’s income was around $84 (highest ever due to some high paying adverts!) and costs were $1.91.

Yes I know it’s wrong, but I wanted to show you that your initial claim on how many paid users you must have to break even was too high.
I Do not know how much your running costs will increase with a certain number of users. But I expect it to be a small percentage of your income if you would in fact sell for $2.
The bigger problem would be getting enough people to buy the game in the first place. I Assume you have written in your terms of use that you do not offer a life time of service. Because that would be impossible due to the costs. Also no user would expect a lifetime client server based game for just $2. Well ok there will be some users that are so out of this world that would expect it, but normal users would be fine with a limited time.

So in short you could say you will guarantee 1 year of service, that would result in a maximum of 365 days running costs without income. This because you are in control when to decide to stop selling the game. The running cost will decrease in the last year so its hard to make a fair estimate on the running cost vs the number of sales. But if you take all of these figures in account it’s still a valid business strategy. I’m not stating its better than free with advertisements. Just that it’s wrong to state you need hundreds of thousands sales to break even. My guess is that with 3 sales a day, for the price of $2 each, you will in fact make a profit over a period of 2 to 3 years. Not a huge profit. But take in account that 3 sales a day is not a high figure to start with.

Now your current model is free + advertisements. Your income per user will be much lower. Normally this model will result in just a couple of cents per user in a whole year. So the profit (income - running cost per user) will be much lower than with a paid model. Of course this is compensated by the fact that the cost per user is lower with much higher users numbers, and the fact that free games tend to be downloaded much much more than paid versions. So this might be the best option for your game.

A combination of paid/free can also be a interesting model.

In short, this whole message comes down to this one fact. Your running cost will not be infinitive if you sell your game, you should only guarantee a certain number of months service.

From David’s latest post:

Total Ad Revenue: $6,855.13
Android Market Sales: $4.04 (2 sales)

I think this gives us an idea of where we should probably focus our time and effort.

Of course, it all depends on the app, many will have more sales and less ad rev, but look at that difference! Wow!!

Paid apps might maximize revenue per-user (or not), but do they maximize revenue per hour spent working on the app?

Why not put a paid app and a free app? Some people just do not like ads and if the like the app they may buy it. Just put it for like 2.99 since I don’t think someone will click on an ad enough times to make about $2.