the facebook measurement divide - Rakuten Marketing Blog

17 downloads 549 Views 3MB Size Report
Aributable revenue from Facebook ad campaigns might be higher than ... discrepancies between Facebook conversion trackin
T H E FA C E B O O K MEASUREMENT DIVIDE

R A K U T E N

M A R K E T I N G

INSIGHTS

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

ME ASUR EM ENT DI S C R E PAN C I E S A!ributable revenue from Facebook ad campaigns might be higher than advertisers' web analytics lead them to believe. Rakuten Marketing Insights analysis of client performance data revealed discrepancies between Facebook conversion tracking and web analytics that are costing advertisers insight into 192% more a!ributable revenue and higher return on ad spend for mobile campaigns. While the discrepancies between Facebook conversion tracking and web analytics for mobile are significant, they are minimal for desktop: There is on average a 3% discrepancy for a!ributable revenue between Facebook conversion tracking and web analytics. Despite the prevailing theory that cross-device – and the inherent challenges associated with cross-device measurement – accounts for this disproportionately high discrepancy on mobile, this is not the case. In fact, the cross-device campaigns analyzed revealed that a!ributable revenue only comprised on average 5.6% of the total revenue generated across mobile-only, desktop-only and cross-device campaigns, and as li!le as 2.4% for one retailer in the study. Advertisers invest their marketing budgets across channels and strategies that yield the highest returns. The discrepancies between Facebook conversion tracking and advertisers’ web analytics are not trivial. For one client included in the study, the a!ributable revenue from Facebook conversion tracking could be as high as $4.7 million annually, while their web analytics would report $1.9 million. Many advertisers apply a 2x – 3x multiplier to their web analytics a!ributable mobile Facebook revenue to estimate a more realistic return on investment for their mobile campaigns. The advertisers relying solely on web analytics are making decisions based on data that doesn’t accurately reflect that Facebook campaigns are driving more revenue than they think. Facebook conversion tracking is inconsistent with what advertisers see in their web analytics for many reasons. Although Facebook conversion tracking paints a more comprehensive picture of performance, there are technological challenges that prevent conversion tracking on Facebook from functioning correctly. This discrepancy means there are additional conversions happening that are simply not recorded anywhere. The issues surrounding measurement are abundant and complex – the following summarizes some of the challenges advertisers face. 2 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

Web Analytics



Measuring post-click and post-impression conversions: Post-click conversions are tracked differently by Facebook conversion tracking than in web analytics: Facebook conversion tracking captures all post-click conversions (we measured post-click conversions within a 24-hour window), while web analytics typically only captures last-click conversions. The same holds true for post-impression conversions: Facebook captures all post-impression conversions within a 24-hour window, and web analytics simply cannot capture these conversions from viewed inventory.



Cross-device consumer identification & measurement: Web analytics platforms that rely on cookies cannot accurately track cross-device conversions because of the inherent challenges with identifying consumers across devices. Coupled with the consumer’s ability to block cookies on desktop and mobile, it becomes nearly impossible to measure conversions and a!ributable revenue for cross-device campaigns. Facebook doesn’t solely rely on cookies and can measure conversions as consumers log into the platform across devices.



Mobile platforms & cookies: Social advertising experts and advertisers agree that some of the discrepancy can be a!ributed to certain mobile operating systems and Internet browser combinations blocking third-party cookies. Depending on how the third-party measurement system is configured, it will appear to the browser as a third-party cookie which will not be tracked, or a first-party cookie which will be tracked.

3 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

MOBILE I S T H E C L E A R W I N N E R FO R PE RFO R M A NCE The other major trend that emerged from this study is that mobile outperforms desktop, and advertisers’ web analytics data is not reflecting this trend accurately. This should come as no surprise – Facebook has 1.57 billion mobile users, and 1.03 billion of them access Facebook daily. Among the Facebook user base, 54% access the platform exclusively from mobile. Mobile advertising has historically experienced challenges, but the writing is on the wall – advertisers cannot shy away from investing in mobile if they want to reach and drive revenue from social-savvy mobile consumers. Analyzing Facebook conversion tracking data made it clear that mobile is the undisputed winner over desktop when evaluating campaign performance. The click-through rate for Facebook mobile ads was on average 63% higher than desktop, although the conversion rate was on average 33% lower. However, when you factor in that cost-per-click for mobile is 70% lower than desktop, return on ad spend for mobile Facebook is 72% higher than desktop. AV E R AG E FAC E B O O K A D ROAS M O B I L E V S D E S K TO P

63% HIGHER

70%

33%

LOWER

LOWER

MOBILE ROAS

72 HIGHER %

THAN DESKTOP

Desktop

Mobile CTR

Desktop

Mobile CVR

4 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

Desktop

Mobile CPC

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

Analyzing Facebook conversion tracking data for each of the individual clients in this study revealed that the same pa!erns from the aggregate data hold true at the individual client level: a clearly demonstrated higher click-through rate, lower conversion rate, much lower cost-per-click and therefore higher return on ad spend for mobile, when compared to desktop. In addition, the reported a!ributable revenue for each advertiser was substantially higher in Facebook conversion tracking than what was reported in each advertiser’s web analytics.

CLIE N T O NE CLIENT 1

Discrepancy: A!ributable revenue on mobile is 58% lower in web analytics than in Facebook conversion tracking, while we saw an a!ributable revenue discrepancy on desktop of 8% between

8%

LOWER

58%



Click-though rate on mobile is 77% higher than desktop.



Conversion rate on mobile is 8% lower than desktop.



Cost-per-click is 68% higher on desktop than on mobile (or 68% lower on mobile than on desktop).



Therefore, mobile achieves a 69% higher return on ad spend than desktop.

Desktop Revenue*

Web Analytics

Facebook

Facebook conversion tracking highlights:

Facebook

Facebook and web analytics.

Web Analytics

LOWER

Mobile Revenue

5 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

C L IE N T T WO CLIENT 2

Discrepancy: A!ributable revenue on mobile is 70% lower in web analytics than in Facebook conversion tracking, while we saw an a!ributable revenue discrepancy on desktop of 8% between

8%

LOWER

Facebook and web analytics.

70%



Conversion rate on mobile is 46% lower than desktop.



Cost-per-click is 87% higher on desktop than on mobile (or 87% lower on mobile than on desktop).

Mobile Revenue

Therefore, mobile achieves a 64% higher return on ad spend than desktop. CLIENT 3

C L IE N T T HR EE

15%

Discrepancy: A!ributable revenue on mobile is 68% lower in web

LOWER

68%

analytics than in Facebook conversion tracking, while a!ributable Facebook

Facebook

tracking than on web analytics.

Web Analytics

LOWER

revenue on desktop is only 15% higher in Facebook conversion

Web Analytics



Desktop Revenue*

Web Analytics

Click-though rate on mobile is 49.4% higher than desktop

Facebook



Web Analytics

Facebook conversion tracking highlights:

Facebook

LOWER

Facebook conversion tracking highlights: •

Click-though rate on mobile is 64 % higher than desktop.



Conversion rate on mobile is 6.6% lower than desktop.



Cost-per-click is 56% higher on desktop than on mobile (or 56% lower on mobile than on desktop).



Therefore, mobile achieves a 122% higher return on ad spend than desktop.

6 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

Desktop Revenue*

Mobile Revenue

* It is important to note that Facebook conversion data is reported with 1-day post-click a!ribution while web analytics report with 30-day last-click a!ribution, which can lead to discrepancies.

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

The data and performance analysis creates some interesting insights: consumers are more likely to click on a Facebook ad on mobile but less likely to convert. Although consumers are converting more and more on their mobile devices, the conversions rates will likely continue to go down, due to the challenge of purchasing on small screens. The substantially lower cost-per-click drives an overall higher return on ad spend on mobile. If the fact that mobile is a critical part of the consumer journey isn’t enough to make advertisers invest heavily in reaching consumers on social media on mobile devices, higher return on ad spend and substantial a!ributable revenue should be. We are not just finding that mobile outperforms desktop – we are finding performance that should make advertisers pause and rethink their strategy when considering the efficacy and budget allocation towards mobile Facebook advertising.

ADV IC E TO A DV ERT I SE R S Consider taking a systematic approach to ascertain how Facebook advertising on mobile is performing, leveraging the following process:

1. CREATE A MULTIPLIER: Measure the difference in mobile revenue between Facebook conversion tracking and your web analytics tool. Facebook Reported Revenue Web Analytics Reported Revenue

We will use $180K as mobile Facebook revenue and $100K as mobile revenue reported by your web analytics tool. $180k/$100k= 1.8 For this example, the multiplier is 1.8. If working with fewer than thirty days of campaign data use an benchmark multiple of (2.5x), a multiplier we recommend to our clients based on working with numerous advertisers and managing their Facebook advertising campaigns. 7 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

2. ACCURATELY ATTRIBUTE MOBILE REVENUE: Multiply the mobile revenue reported by your web analytics tool by the multiplier created to get your net a!ributable mobile revenue. Mobile Revenue Reported by Web Analytics Tool

x

=

Net A!ributable Mobile Revenue

$100k x 1.8 = $180k This number matches the original Facebook reported mobile revenue to show the multiplier bridges that gap.

3. DE-DUPE THE REVENUE USING RELIABLE DESKTOP DISCREPANCY: Facebook measures performance using a post-click revenue model and web analytics tools typically measure a!ributable revenue based on last-click performance. This discrepancy must be accounted for by normalizing the data. This can be done by using the revenue discrepancy between Facebook a!ributable revenue on desktop and that of the marketer’s a!ribution so$ware as a benchmark. For this example, we will use $120k in desktop Facebook revenue reported and $100k in desktop revenue reported by web analytics. Desktop Facebook revenue reported

-

Desktop Revenue Reported by Web Analytics Tool

Desktop Revenue Reported by Web Analytics Tool

=

Desktop Discrepancy

Net A!ributable Mobile Revenue - ( Net A!ributable Mobile Revenue x Desktop Discrepancy )

=

Accurate Mobile Revenue

($120k - $100k)/$100k = 20%. Then discount the mobile multiplier revenue by that desktop discrepancy. $180k – ($180k * 20%) = $144k. 8 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED

T H E

FAC E B O O K

MEASUREMENT

DIV IDE

METHOD O LO GY: Rakuten Marketing Insights analysis of client performance across Facebook advertising on desktop and mobile for a one-month period. Data analyzed was from Facebook conversion tracking and advertisers’ web analytics reporting across Adobe Analytics, Coremetrics and Google Analytics. The three clients included in the study spend between $350,000 and $725,000 annually to advertise with Facebook.

S O U RC E S: h!ps://www.facebook.com/business/help/community/question/?id=10208936989490685 h!ps://searchenginewatch.com/2016/02/16/11-reasons-why-your-facebook-ad-measurement-is-messed-up/ h!ps://www.facebook.com/help/community/question/?id=10153294676310223 h!p://socialmediatipz.com/facebook-clicks-dont-match-google-analytics/

CON TACT U S Rakuten Marketing www.rakutenmarketing.com [email protected] 800.519.1432

9 © 2016 RAKUTEN MARKETING, LLC. ALL RIGHTS RESERVED