Facebook’s policy changes regarding virality keep changing the social gaming industry. Notifications channels have been significantly tightened in favor of a better user experience, ending the honeymoon phase in which social game developers were able to grow at a tremendous pace for free. Today, it has become increasingly harder for new players on the market to spread their games on Facebook because of higher competition and budget restraints. As a new Facebook game developer you are basically left with two choices: to be convinced that a really good game will still spread virally or to invest in ads. You can run the ads yourself or hire the services of specialized companies. Many developers, however, are still skeptic about “purchased users”. We talked to Ali Merali from AdParlor, a large Facebook campaign leader, about the costs, quality and sustainability of purchased Facebook users:
SocialGamesObserver: Why should I pay AdParlor and not generate my own Facebook Installs?
Ali Merali: Application developers on the Facebook platform, who have a user acquisition strategy in place, will know that purchasing ads to promote their application can only be done on a CPC or CPM model. With privileged access to the Facebook Ads API, AdParlor’s Optimization Engine eliminates the ‘guess work’ with CPC ad buying, to provide a performance based CPI/CPA purchasing model. This means that you will only be charged for new and unique users that are brought to your game. On a higher level, AdParlor’s engine optimizes each individual ad across the many targetable variables (age, gender, location, ad copy etc…) with automatic bid adjustments every few minutes, ensuring the highest volume of users at the most competitive prices on the platform.
SGO: How much does an acquisition cost? Can you explain the pricing?
Ali: The AdParlor pricing model remains consistent with any true bidding system- that is, you define your CPI rate and we will generate as much volume as possible at that given rate. The main factors that influence pricing (in no particular order)
- Conversion Rates- From click of ad, to install of application (standard: 45%-60%)
- Click Through Rate- CTR expressed as a percentage of clicks per every thousand impressions (standard: 0.06%-0.250%)
- Application Appeal- The CPI for a casual game is much lower than the CPI for a hardcore MMO game.
- Application Saturation- If the game has already reached hundreds of thousands of players, the next user becomes more expensive than the last (more on this topic below)
- Target Market- The CPI of a US male between 25-29, will be 5x expensive than a 25-29 male from Malaysia. You can view an entire list of CPI estimates, by country, in a recently published white-paper – http://www.adparlor.com/FBInstallWhitepaper.aspx
As an example, a well appealing game with 300,000 MAU’s and ads that can sustain 0.1% CTR and 50% Conversion rate, should be able to generate 5000 installs/day at a $0.50 CPI (targeting prime English speaking countries)
SGO: How does the recent change in Facebooks notification policy change your services?
Ali: As we all know, Facebook is always changing their policies and it is upon us to both anticipate change and of course easily adapt. Most people would think that with notifications turned off, along with other viral channels continuously clamped down on, application developers would have no other choice but to increase their paid user acquisition efforts to grow their game. Although true, we do end up seeing that smaller (and often less funded) developers, find more difficulties in producing an ROI positive Facebook game. With this in mind, AdParlor has developed more features to steer user acquisition budgets more intelligently, ensuring the most qualified users are being sent.
SGO: Can you elaborate on the marginal increase in costs as MAU’s climb?
Essentially, when you launch your game and begin a very targeted user acquisition campaign, ads for your game are to be shown to users who are most susceptible to click, install and engage with your game. Click-through-rates and conversion rates are relatively high and hence CPI’s can be set low, while achieving desirable volume. However, as this initial user base is exhausted, you have two choices to sustain prior volume:
1. Broaden targeting options (age, gender, location, keywords etc…). This may result in lower quality users
2. Increase CPI such that Facebook continues serving the ads for your game to your most targeted market
Most developers will find a comfortable mix between the two, such that their life-time-values (LTV) remain above their CPI costs.
SGO: What is the average launch budget to get 200.000 MAUs?
Ali: This is a difficult estimate to make, breaking down the estimate by game category can help the approach:
Casual- Strategy & Games of Chance (Bingo, Mahjong, Uno…): $60,000-$70,000
Casual- Virtual World (Harvest, Manage, City Building…): $80,000-$100,000
RPG- (Hero’s, Mafia, Crime…): $90,000-$110,000
Hardcore MMO- (Kingdoms, Warriors, Vikings…): $115,000-$130,000
The above estimates reflect budgets for paid users. When factoring virality, you can reduce the budgeted numbers by 1/3 or even ½ , in determining your costs to reach 200,000 MAU’s,
SGO: Which countries monetize best?
Ali: ‘Prime English’ countries, the United States, Australia, United Kingdom and Canada produce the highest monetizing users. As Facebook continues its global penetration, we see a surge in promotions for European traffic (Denmark, Germany, Netherlands, Sweden etc…) which have proven to also monetize well. Moving away from these markets involves strategic local partnerships to ensure your traffic is being monetized effectively. It will be interesting to see if Facebook can further penetrate the Japanese market, since those users commonly produce the highest ARPU’s. If so, I think we’ll see a wave of localization from Western developers to the Japanese market.
SGO: Is it really a sustainable business model for smaller developers to acquire traffic through CPA deals?
Ali: It certainly is. So long as the developers have either built or are using a strong analytics platform to intelligently influence their acquisition campaigns. For small developers, we often recommend spending lesser amounts in their first week or two as they test their game mechanics and can easily make adjustments and tweaks. While doing so, they should be analyzing which segments of traffic are performing the best for them.
Ex. males, 18-21 from Austria are highly engaged, viral and monetizing… shift more of your ad spend to this segment while still fishing around for other gems in the many targeted combinations.
SGO: How does bought traffic compare to viral traffic?
Ali: It’s quite normal when purchasing traffic from facebook right-hand column ads, that an equal or slightly lesser amount of viral users additionally join into the game being promoted. These viral users tend to perform very well in terms of retention and engagement. With this in mind, developers must allocate marketing budgets to seed their applications, but more importantly to maximize the viral growth potentials.