Previously, we published  an article looking at average revenues per paying user by age clusters with blogger and entrepreneur Andrew Chen.

It’s always good fun to look at data, but given these revenue discrepancies between older and younger users, what’s a developer to do?

This week, we followed up with 3 tips for monetizing social game users.

From the full article:

Get Them Young: 3 tips to monetize younger users

1. Think volume. Look for the users who are transacting the most, and then make sure you understand exactly who they are (and how they might be changing). For example, today your revenue may be driven by a massive group of teenagers, but what will happen when those teens become 20-somethings? In this series, we explored this question by age, but you’ll also want to think about geography, language, and gender. ‘Think volume’ means:

• Mind your game. If your product is subpar, you shouldn’t expect amazing volumes or revenues, no matter how much you…
• Focus on growing traffic through virality. How can you make your game even more social, more addictive, and more spreadable?
• Get users to complete. Users are 3 times as likely to make additonal payments if they’ve completed at least one offer.

2. Hold on to your users. People of all ages get tired of games easily. The last thing you need is a poor user experience to push users over the edge and straight into the database of a competitor. Do certain offers just rankle your userbase (leading to poor conversions, bountiful complaints, and churn)? While your payments solution’s algorithms will help you find the best offers for your users, there are always going to be a couple that just don’t perform. ‘Hold on to your users’ means:

• Pick out and remove underperforming offers, either individually or by offer category, and address customer complaints. For example, ‘adult’ offers may not work well if your game’s users are primarily 13-17 year olds.
• Diversify your product(s). How can you enrich a single game to be more complex and engaging? How can you offer more complementary games so when a user defects, she defects to another game in your suite?

3. Keep your eye on empty spaces. Yes, Facebook is huge. Yes, Zynga is dominating. But, growth potential is everywhere still. As more users of all ages sign up for their first Facebook accounts, more people pour into the virtual economy. As Facebook grows in locales outside the U.S., so do the games and apps that inhabit its ecosystem. As users get tired of specific games, they’ll start looking for other places to spend their time and money. They’ll probably invite their friends, too. ‘Keep your eye on empty spaces’ means:

• Don’t make a play just because someone else is making bank off of it (for now). Today’s leaders got there because they kept their eyes on empty spaces and filled them, quickly.
• Look for under-monetized user groups. How well is your game doing with young males? Can you work in a way for more of these users to complete their first offer (and open the door to additional payments)?

These should be your main considerations:

Growth
What does the growth trajectory look like for young users? How many of these users are already playing games, and how many more aren’t? The online casual games industry is still young and has plenty of room for growth.

Facebook boasts 300 million active users, with almost a third of these in the U.S. Since the entire population of the United States is just over 300 million, that means approximately half of all U.S. internet users, or a third of the entire U.S. population, are on Facebook.* Facebook counts 70% of users as having ‘engaged with a Platform application,’ meaning that most users have loaded an app of some sort at some point in their Facebook time. Judging by the impressive monthly active uniques the biggest developers are enjoying (51MM for Zynga’s Farmville alone), it seems that games have already taken off on the network. With all this, is there still room to grow?

Yes. Here’s why:

• Facebook has saturated the U.S. market, but that doesn’t mean every Facebook user is playing a game. Yet.
• The U.S. isn’t the only country in the world, either. In terms of Facebook traffic growth rates, the U.S. doesn’t even make it into the top 10. As other economies (real and virtual) catch up, markets around the world should start looking more and more promising for developers looking to monetize.
• People get tired of games. One developer’s churn is another developer’s new user.

As mentioned above, younger users contribute the lion’s share of total revenue for virtual transactions – for now. However, Facebook reports that the 35 and up group is their fastest growing demographic, so will we see this shift reflected in game usage and monetization too? Probably. But until the older users reach critical mass on the network, would you rather be competing hard for the same handful of housewives or slyly going for the many younger users at lower ARPPUs and massively higher transaction volumes?

Changing ARPPUs
Do ARPPUs change as users get older? Will your 15 year old user be worth more after she turns 18, gets a better job, and starts opting for direct payment over offers? We know that the typical 18 year old makes you more money than the typical 15 year old, so from this we might guess that it will pay off to hold onto that user as she ages.

Age     ARPPU
15     $2.65
18     $2.92
22     $2.82
25     $2.99
29     $3.33

Older users
Should you try to grow your older userbase? As just mentioned, Facebook’s fastest-growing demographic is the 35 and up set. While actively trying to acquire these users (over others) may divert your resources in ways you can’t afford, it’s likely that your game will indirectly absorb the benefits of Facebook’s demographic growth anyway. If everyone else is focusing on winning the middle-aged housewife segment, would you be better off stealthily (and expertly) acquiring the forgotten younger users? Try it. Measure it. Report back.

Conclusion

In parting, don’t buy into a ‘must do’ (eg. housewives) just because it’s popular today. Popularity doesn’t mean it’s wrong, but it does probably mean that lots of other developers are out there thinking the same thing as you. Instead, look at the data and focus your work where the greatest opportunity currently blossoms. Right now, that’s users who are in their teens and mid-20s.

If you’ve been targeting and you’re seeing interesting results, please share in the comments. What’s worked for you, and what would you do if you were a new developer just entering the marketing today?

One of the drawbacks to this system, however, is that as developers focus on how to make that high-level tier of virtual goods and its correlative currency as desirable to the users as possible, the basic, or free, currency often gets ignored or forgotten.

The most common way in which basic currency becomes worthless is irrelevance: When all the virtual goods that hold value to a user are only purchasable using premium currency, the basic currency becomes irrelevant to the user. In Habbo.com, for example, pixels (the basic currency) can only be spent on a disappointing fraction of the virtual goods available in the game, and even of those, most pixel purchases required a co-payment of coins (the premium currency). Pixels alone are subsequently worth next to nothing in Habbo– an unfortunate consequence of putting premium currency on a pedestal.

For a developer who wants to maximize numbers of paying users, it seems logical. Why should a game be designed to allow the basic currency to retain value when users get it for free? The premium currency, which (in the absence of a subscription model) is the sole provider of profit for the developer, should be favored and exalted, because the more worth it holds in the game, the more users will want and pay for it.

The error in this logic, however, is the failure to remember the age-old adage: Time is money. Users may not be spending money in order to get basic currency, but in most games, they pay for it with time: time spent completing challenges or tasks, time spent signing into the game and visiting other players– time spent, in other words, simply playing the game.

In DogWorld, users can earn basic currency just by logging in.

Basic currency is a reward for user activity, and when that currency can be exchanged for virtual goods that have purpose, users will continue to desire that currency and, likewise, will continue to be active. And any successful social game or community developer can assure you that retention of a loyal, active and interested player is worth far more than a one-time $19.99 purchase of premium currency.

In other words, as a developer, you shouldn’t overlook what your basic currency can do for your game just because it isn’t a direct transaction into real profit. In the long run, its indirect ability to retain and engage your users will benefit your game as a whole– including revenue– so long as you allow it to stay valuable in comparison to your premium currency.

One easy suggestion? Don’t limit the virtual goods payable by basic currency to the lower tiers. Just as exclusive goods can encourage users to spend money in pursuit of the status symbol, so can they encourage gameplay for that same reason. Providing valuable virtual goods that must be obtained with basic currency means users cannot simply buy their way through the game, but instead must put in the hours.

Remember: successful free-to-play games don’t require a user to spend money in order to enjoy it, either on a short-term or long-term basis; spending money should only enhance an already-solid user experience.

What could have possibly gone wrong?

CPM factors: A/S/L?

If your CPM is a number that tells you how much revenue you’re generating per so many users, it should come as no surprise that CPMs are affected by (hey!) your users– namely, who your users are and in what part of the world they live.

• Age: How old are your users? Are they mostly kids under 18 or middle-aged 9-to-5-ers? Younger demographics are less likely to have their own credit cards and will often require parental consent for a legitimate transaction to go through, so if this is to whom your game predominately caters, daily CPMs could be lower than your game model might otherwise predict. As the user age goes up, so does the likelihood of users buying, with those in the 25-40 range being the most willing to spend.
• Sex: In the world of stereotypes, gamers are male, people who spend a significant amount of time on the computer and online are male, so correspondingly, online gamers must be male. In the world of reality, however, females steadfastly hold an equal share in the casual-gameplay market, and according to Tina Tran of Girls In Tech, women are “making up 75% of the population that pays to play.”
   
  Does this mean that games whose users are mostly male won’t make any money? No; it just suggests that the more women (or, taking the above into account: the more older women) a user base contains, the higher the percentage of paying users will be, leading to increased CPM.
• Location: You feel like your game could have a global appeal, so you’ve gone ahead and translated it into 15 different languages (with 20 more to be added soon), and it’s worked! Your international traffic is skyrocketing– but now your CPMs are plummeting. Did you remember to make sure that as your game added support for more and more countries, your payment options followed suit?
   
  Your users might be coming from the 190 markets that Paypal services, but Paypal might not be an appropriate service for them, as it requires either a bank account or credit card. If you provide a mobile payments option, do the countries your mobile payments provider supports match the countries from which you’re seeing a high volume of traffic? If you provide alternative payments, are there a lot of offers available for those countries? Your users may be in a position where they’d be more than happy to give you their money, but they have no way to do so, and this will effectively bring down your CPMs.

Other factors that affect CPM

Timing is everything

When was the last time you checked your stats? Five hours ago? Five minutes ago? Maybe you should check them right now. Are numbers up? (Rejoice!) Are they down? (Despair!) Wait. Check again– anything new now?

CPMs– like the stock market, foreign exchange rates, blood pressure, or even brainwave activity– are not static throughout the course of any given day, and if you obsess over your real-time numbers and react to every fluctuation, you’re bound to be in for an emotional roller-coaster ride. Dramatic changes are worth concern, of course, but for the most part, CPM will naturally rise and fall throughout the day, particularly if you have a widely-varied user base.

If most of your users are still in school, you can’t expect your reports to show much at 11 a.m. on a weekday. Similarly, if most of your paying traffic comes from Germany, expect numbers to be lower when you check stats at 7 p.m. Pacific time; that might be a prime revenue-generating hour for the U.S., but where your users are, they’re still fast asleep.

Advertisers make mistakes, too

Alternative payments aren’t a flawless system, and so it happens that sometimes a user will complete an offer but not receive credit from the advertiser associated with that offer for doing so. Until this discrepancy is resolved, then, CPMs are artificially lowered, as the user has now generated revenue that has yet to be credited to the game.

In conclusion

If you use CPMs to measure your game’s monetary success, take some time to look into the comprehensive factors behind the numbers instead of just taking them at face value.

• You can tailor a game to perfection, but if you fail to pay attention to who’s playing your game, your CPMs are bound to stall at some point. Make sure that your payment options support your users. If they’re young, consider mobile payments (as younger users are more and more likely to have a cell phone these days) or age-appropriate offers that don’t require a credit card. Or, if they’re in foreign countries, look into payment providers that work in global markets but don’t require a credit card and consider showing more offers that take that international traffic.
• Instead of thin-slicing, look at the bigger picture. Don’t make comparisons in numbers hour-to-hour, but rather day-to-day, or even week-to-week (as weekday performance is commonly better than weekend performance). Then, taking into account any changes to the game’s development or user base, you can form your conclusions and act accordingly.

(Ever wonder, by the way, why the abbreviation is “CPM” when “thousand” begins with a “T”? “M” is the Roman numeral for 1000, as in Latin, “thousand” = “mille“.)

There’s a debate over whether CPM is really the best evaluative metric, but in any case, as long as it’s the standard number used to measure success, it’s good to have a better understanding of the many factors that affect CPM. There are two categories of factors– those controllable by the developer and those not. For now, let’s just look at the first category, which includes demand for currency, engagement of the game, and retention of the game.

Demand for currency

Is the game designed so that users want or see the need for purchasing currency (and again, “currency” and “virtual goods” are interchangeable as terms for the end goal)? Last week, it was discussed how virtual goods need to provide value to the user’s experience in order to be in demand– the more valuable those goods are to the game, the more currency users will purchase in order to obtain those goods, and the more revenue a game generates, the higher the CPM will be.

Engagement of the game

How long are users staying on the game each time they visit the site or application? The more time users spend immersed in the game environment, the more active they are and the more likely they will be to visit the payments page and participate in a monetary transaction (either through direct payments or offer completions). By designing a game or community that really appeals to its users, those users become more willing to invest in that game. No one wants to spend money on something they have only mediocre feelings about.

Retention of the game

Do users consistently come back, or do they play once or twice and never return? Similar to above, a game that holds more long-term users will have higher CPMs, since long-term users will be more invested– either in time or money, or both. It’s a self-perpetuating cycle: the more a user plays, the more likely that user will be to purchase currency to progress in the game, and the farther along in a game the user is, the more likely he or she will be to return and keep playing.

Because it generally takes new users a certain amount of game play before they can see the value or feel the need of purchasing currency, a game that always has only new users will never see the consistent stream of revenue activity that longer-term users would produce.

**********

These are all elements directly under the developer’s control that can be adjusted to ensure that CPMs are maximized: Make your currency valuable so that users are always willing to buy more (a user who purchases currency only to find very little benefit to having done so will not be likely to purchase currency again in the future). Make your game engaging so that users like the game enough to not mind putting money into it. Focus your game strategy on long-term play so that your users who have finally reached the stage where they see the value of buying will keep returning.

The other major factors that affect CPMs include a game’s demographics (in particular user age groups, gender, and country) and, if offers are used as an alternative payments option, advertiser errors. Look for next week’s analysis piece to see the breakdown of these factors and what developers can do to best handle them.

The infamous, short-lived "I Am Rich" iPhone app did nothing other than cost $1000, leading six people to willingly buy it for the exclusivity and status.

Inside of a game, that mentality doesn’t change. Whether we’re purchasing real, tangible objects or virtual, pixellated images, the driving question behind whether or not we pull out our wallet is the same– what will this do for me?

The general purpose of virtual goods (VG) is to make game play more interesting, more engaging, more enjoyable– and also to make game developers more revenue. In the case of casual social game developers (particularly those on Myspace and Facebook), VG are their only driving components for revenue, with the following understandings:

• Typically, users exchange real money for the game’s virtual currency, then use that currency to purchase VG– however, the VG are the end goal, so by the transitive property, users are paying for VG

• For the sake of simplicity, VG refers to both objects that can be obtained and actions that can be performed within the game

But even for games that have an initial upfront cost (like the Sims) or that require a paid subscription (like WoW), VG can still be a solid source of revenue. Either way, if developers want to make any profit from VG sales, they need to really understand the dynamics of their game and tailor their VG appropriately: users will not want to pay for virtual goods that do nothing to enhance their game play experience.

How can VG enhance game play?

• Progression: owning the item allows users to increase skill points, which assists in leveling up (e.g. better weapons that enable more wins)

• Amusement: owning the item makes the user happy (e.g. unique-print wallpapers or personality-reflecting furniture for a user’s virtual room)

• Status: owning the item allows the user to show off either wealth or devotion to game play, or both (e.g. highly expensive or limited-supply objects)

Examples of virtual goods in Extreme Basketball that enhance game play by increasing the user's odds of winning matches.

And for VG as actions, enhancement comes predominately via progression and amusement.

So if you’re a game developer, when deciding what sort of VG you’re going to make available to your users, ask yourself: why would my users want this? How does this add worth to my game for them, and how does it make their experience more enjoyable? Once you make VG valuable and make that value evident to your users, the purchases will happen and your percentage of paying users will grow.

Landing page for a poll-submit offer.

It’s true that instead of paying with money, a lot of “free” offers will require the user to pay with their time, their personal information, or both. Common free offers include poll-submits (e.g. “Home Depot or Lowes?” or “Who makes the best pizza?”), which require submission of a poll answer, the user’s e-mail address, mailing address and telephone number (which, because no one ever reads the Terms of Service or Privacy Policy statements, users consent to have shared with third-parties upon submitting)– and in some instances, the incredibly tedious task of having to click through what feels like an endless succession of pages filled with other offers that the user may choose to complete or skip over before ultimately reaching the final page, whereupon the virtual currency is awarded. Quizzes and “gifts” (”Get two free tickets for Southwest!”), other typical forms of free offers, operate in this same way (for quizzes, the final page is also where the user’s results are shown).

Likewise, there are free offers to join survey panels: the user signs up as a new member and usually has to complete at least one survey before getting credited for the offer. Personal information has to be submitted during registration, of course, but it’s not at stake of being shared, so survey panel offers really only demand a user’s time and opinions as payment.

Finally, however, there are the free offers that require nothing from the user other than the willingness to try out a product, whether it be a browser toolbar, an IM client, a Facebook app, an online game or community, or even credit cards or insurance programs. These are the offers that are as free as free gets, and while the payouts rarely come close to those of poll or quiz offers, there is no loss of privacy or time in the completion of these offers.

Some examples of popular free offers: an application for a credit card, a Starbucks card gift offer, and a toolbar download.

And why do free offers matter? Because for a developer of a game or community that utilizes virtual currency, free offers can generate up to 26% of that developer’s revenue– but what’s more, free offers can be a great way to transition more users into directly purchasing that currency. As users get into the habit of doing something external from gameplay in order to obtain additional currency and are able to see or experience the value of having that surplus, they become far more likely to eventually turn into a paying user.

Do free offers come at a cost? Sometimes, yes, as evidenced above; but sometimes, that free offer that promises to only take a moment of your time really does take only that. And to the developer who wants to turn more users who are hesitant to spend even a dime on virtual currency into users who want to pay, free offers are better than free– they’re the ticket to increasing revenue.

[1] It’s worth noting that the requirements to complete the offer and receive currency are completely different from the requirements to receive the featured gift item.