devices. Ads in News Feed have a significantly higher average price per ad due to factors which include the prominent position of the ads. The increase in price per ad in the United States was partially offset by an increased percentage of our worldwide ads being delivered in the Asia and Rest of World geographies where the average price per ad, while growing on a year-over-year basis, is relatively lower. The average price per ad was also affected by a decline in the average price per ad in Europe in 2012 compared to 2011 due to the impact of foreign exchange rate changes, an increase in the percentage of ads being delivered in European regions where the average price per ads is relatively lower, and in part, we believe, to continuing weak economic conditions in that region affecting advertiser demand. 2012, comparisons to prior year are not meaningful. The increase in ads delivered was driven by user growth and certain product changes, including the addition of News Feed ads on personal computers and mobile devices. MAUs grew 25% from December 31, 2011 to December 31, 2012 and average DAUs grew 28% from December 2011 to December 2012. Additionally, in the fourth quarter of 2012, we lowered our market reserve price (i.e. the minimum price threshold accepted in our ads auction), and this product change had the effect of increasing the number of ads delivered and decreasing the average price per ad. This change primarily affected the Rest of World and Asia markets where the average price per ad is relatively lower, and the change increased the percentage of our ads that are shown in relatively lower priced markets, which has the effect of decreasing the overall average price per ad. For the fourth quarter of 2012, we estimate that mobile advertising revenue as a percentage of advertising revenue was approximately 23%. compared to 2011. Facebook Payments became mandatory for all game developers accepting payments on the Facebook Platform with limited exceptions on July 1, 2011. Accordingly, comparisons of Payments and other fees revenue to periods before this date may not be meaningful. 2012, we had deferred recognition of Payments revenue until the expiration of this period as we were unable to make reasonable and reliable estimates of future refunds or chargebacks arising during this claim period, due to lack of historical transactional information. Beginning in the fourth quarter of 2012, we had 24 months of historical transactional information which enabled us to estimate future refunds and chargebacks. Accordingly, in the fourth quarter of 2012, we recorded all Payments revenues at the time of the purchase of the related virtual or digital goods, net of estimated refunds or chargebacks. This change resulted in a one-time increase in Payments revenue in the fourth quarter of 2012 of approximately $66 million as we recognized revenue from four months of transactions. respectively, came from a single customer, Zynga. Revenue from Zynga consisted of payments processing fees related to their sale of virtual goods and from direct advertising purchased by Zynga. |