Operational Attribution with Google Analytics
11 likes2,661 views
The document discusses the evolving concept of conversion attribution in digital marketing, emphasizing the importance of recognizing all interactions that contribute to conversions instead of relying solely on the outdated last-click model. It proposes a simplified framework called 'see-think-do' to help marketers adapt their strategies and metrics for better decision-making and budget allocation. Additionally, the document stresses the need to use complementary data and advanced modeling techniques to improve understanding and application of attribution across various marketing channels.
![OFFLINE CONVERSION MEASUREMENT
USING THE MEASUREMENT PROTOCOL
Code to extract GA’s CID of a user for later use in Measurement Protocol
<?php
function gaParseCookie() {
if (isset($_COOKIE['_ga'])) {
list($version,$domainDepth, $cid1, $cid2) = split('[.]', $_COOKIE
["_ga"],4);
$contents = array('version' => $version, 'domainDepth' =>
$domainDepth, 'cid' => $cid1.'.'.$cid2);
$cid = $contents['cid'];
}
else $cid = gaGenUUID();
return $cid;
}
?>
NON-GOOGLE CODE
use at risk
your own](https://image.slidesharecdn.com/gaucwarsaw-attributionw-gap-25september2014-noanimationversion-v2-140926070420-phpapp01/85/Operational-Attribution-with-Google-Analytics-61-320.jpg)
Operational Attribution with Google Analytics
- 1. OPERATIONAL ATTRIBUTION a framework and a methodology to make actionable attribution in GA Jonathan Breton, Google GAUC WARSAW Sept 25th
- 4. Source: eMarketer October 2013 $
- 5. DEFINITION OF ATTRIBUTION IS EVOLVING AS WE SPEAK “Conversion attribution is the practice advertisers use to give appropriate credit to every impression, interaction or click that helped drive a conversion in a campaign” % % % % %
- 6. DEFINITION OF ATTRIBUTION IS EVOLVING + impact of offline marketing “Conversion attribution is the practice advertisers use to give appropriate credit to every impression, interaction or click that helped drive a conversion in a campaign” efforts +predictive modeling + cross-device tracking + offline value of a conversion + audience management AS WE SPEAK + distinction New vs Loyal cust. +actionability
- 7. How is Attribution doing today? How can we use it to change the way we think about media? How can I find insights in Google Analytics and start experimenting? What’s next after online attribution?
- 9. ADVERTISERS TRACK DATA THAT THEY DON’T USE last-click attribution considers only 15% of the data you have collected source: Average of Google Analytics premium experiments in SEEMEA ?
- 10. MOST AGREE THAT ATTRIBUTION ENABLES BETTER BUDGET DECISIONS 72% MOST USE THE LAST-CLICK MODEL TO MAKE THEIR BUDGET DECISIONS 79%
- 11. THE LAST CLICK MODEL IS A DECADE OLD CONSENSUS last convert ? ? ? ? ? ? ? ? click
- 12. MOST MARKETERS STILL OPERATE IN A SILOED WORLD SEO PAID SEARCH DISPLAY SOCIAL EMAIL MEDIA
- 13. MARKETERS NEED GUIDANCE AND METHOD “Need to better understand the potential” 42% “Lack of priority within marketing” 51% source: Marketing Attribution Management 2013 Buyer's Guide “Unsure how to choose a method/model” 42%
- 14. TODAY’s DUTY : To give you simplified framework and a methodology to crack the attribution discussion in no time
- 15. See - Think - Do a simplified framework
- 16. Avinash Kaushik Google Analytics Evangelist & Market Motive Co-Founder
- 17. ALL AUDIENCES ARE NOT EQUAL AT A GIVEN TIME, THEY ARE IN DIFFERENT CONSIDERATION STAGES Men who wear shoes Men who are thinking they need new shoes for the fall Men who are looking to buy fall shoes now A D VO C A C Y & L O Y A L T Y SEE THINK DO A W A R E N E S S C O N S I D E R A T I O N P U R C H A S E E V A L U A T I O N
- 18. STANDARD ASSUMPTIONS OF MEDIA CERTAIN MEDIA REACHES CERTAIN AUDIENCES S E O S E O D I S P L A Y A F F I L I A T E S S E M SEE THINK DO
- 19. BREAKING DOWN ASSUMPTIONS RE-ALIGN MARKETING FOCUS WITH CONSIDERATION PHASE S O C I A L S O C IA L S E O S E O S E O A F F I L I A T E S D I S P L A Y D I S P L A Y D I S P L A Y Y O U T U B E Y O U T U B E Y T S E M S E M S E M E M A I L E M A I L SEE THINK DO
- 20. PAID MEDIA PLAYS MANY ROLES WIDE SCOPE OF TARGETING & OBJECTIVES DEMO / PSYCHO / SOCIAL IN MARKET / SIMILAR AUDIENCES / TOPICS INTERESTS / AFFINITY / REMARKETING REMARKETING / CONTEXTUAL PRODUCT-DRIVEN / OFFERS YOUTUBE MASTHEAD / FIRST WATCH DEMO/PSYCHO/SOCIAL/INTEREST CATEGORY IN MARKET / SIMILAR AUDIENCES / TOPICS INTERESTS / AFFINITY / REMARKETING PRODUCT VIDEO BRAND / PRODUCT LOWER FUNNEL INDUSTRY GENERIC YOUTUBE SEARCH / CATEGORY GENERIC / MID FUNNEL SEE THINK DO
- 21. WHY DO WE STILL USE LAST CLICK? CAMPAIGNS MAY NOT GENERATE (ANY) CONVERSIONS RIGHT AWAY Conversions STANDARD CAMPAIGN METRICS OF SUCCESS Conversions DEMO/PSYCHO/SOCIAL I N M A R K E T / S I M I L A R A U D I E N C E S Conversions R E M A R K E T I N G / C O N T E X T U A L P R O D U C T - C E N T R I C / O F F E R S CPA ROI CPA ROI Y O U T U B E M A S T H E A D / ? ? P R O D U C T D E M O /P S Y C H O / S O C I A L V I D E O CPA ROI B R A N D / P R O D U C T / L O W E R F U N N E L T E R M S G E N E R A L B R O A D Y O U T U B E S E A R C H / C A T E G O R Y / M I D F U N N E L / P R O B L E M S O L U T I O N SEE THINK DO
- 22. LET’S MEASURE BETTER METRICS SHOULD ALIGN WITH MARKETING OBJECTIVES REVISED CAMPAIGN METRICS OF SUCCESS (EXAMPLE) ● CTR ● Page Depth ● Store Locator views ● Micro-conversions ● Assist Conversions & ● Impressions/Reach ● % of New Visits ● Bounce rate ● Assist Conversions & DEMO/PSYCHO/SOCIAL I N M A R K E T / S I M I L A R A U D I E N C E S ● Visitor Loyalty ● Abandonment Rate ● Conversions/CPA ● Revenue/ROAS ● Profit R E M A R K E T I N G / C O N T E X T U A L P R O D U C T - C E N T R I C / O F F E R S Ratios (over long-term) Y O U T U B E M A S T H E A D / ? ? P R O D U C T D E M O /P S Y C H O / S O C I A L V I D E O B R A N D / P R O D U C T / L O W E R F U N N E L T E R M S G E N E R A L B R O A D Ratios (over mid-term) Y O U T U B E S E A R C H / C A T E G O R Y / M I D F U N N E L / P R O B L E M S O L U T I O N SEE THINK DO
- 23. WHAT MARKETERS CAN ACCOMPLISH TODAY WITH GOOGLE ANALYTICS
- 24. DIFFERENT VIEWPOINTS ONLINE-TO-OFFLINE CROSS-DEVICE INTRA-CHANNEL Conversion Generic Search Brand Search INCREMENTAL CUSTOMER INSIGHTS MULTI-CHANNEL Display Conversion Mobile Search Conversion Desktop Search Desktop Search Store Visit Conversion Search
- 25. DIFFERENT VIEWPOINTS INCREMENTAL CUSTOMER INSIGHTS INTRA-CHANNEL MULTI-CHANNEL Display Search Conversion ONLINE-TO-OFFLINE Desktop Search Store Visit Conversion CROSS-DEVICE Mobile Search Conversion Desktop Search Generic Search Brand Search Conversion
- 26. A STEP BY STEP PROCESS 1 - measure the potential of attribution 2 - design a See-Think-Do channel grouping 3 - adapt your KPIs 4 - apply relevant models
- 27. MEASURE WITH GOOGLE ANALYTICS MULTI-CHANNEL FUNNELS REPORTING Overview Assisted Conversions Top Conversion Paths Time Lag Path Length
- 28. A STEP BY STEP PROCESS 1 - measure the potential of attribution
- 29. A STEP BY STEP PROCESS 1 - measure the potential impact of attribution on your mix-media Look at the size of your “assisted” online business
- 30. WHAT IS AN ASSIST INTERACTION? Generic Search Click Referral Brand Search Click Display Impression Assist interactions Remarketing Click Last interaction Purchase Product Search Click
- 31. A STEP BY STEP PROCESS 1 - measure the potential impact of attribution on your mix-media 70% of conversions are assisted
- 32. A STEP BY STEP PROCESS 1 - measure the potential impact of attribution on your mix-media 80% of conversion paths contain at least 2 different channels
- 33. A STEP BY STEP PROCESS 1 - measure the potential impact of attribution on your mix-media 79% of your revenue are impacted by Attribution-driven decisions
- 34. A STEP BY STEP PROCESS 1 - measure the potential impact attribution on your mix-media tools used: 1 - USER SEGMENT “ASSISTED CONVERSION USERS” 2- ATTRIBUTION OVERVIEW 15%* ROI increase on assisted conversion = 12% overall business improvement ! *: realistic estimate of ROI improvement based on existing case studies
- 35. A STEP BY STEP PROCESS 1 - measure the potential of attribution 2 - design a See-Think-Do channel grouping
- 36. A STEP BY STEP PROCESS 2 - Assign your channels and campaigns to See-Think-Do roles Redefined channels Sample data for demonstration purposes only
- 37. A STEP BY STEP PROCESS 2 - Assign your channels and campaigns to See-Think-Do roles Sample data for demonstration purposes only
- 38. A STEP BY STEP PROCESS 2 - Assign your channels and campaigns to See-Think-Do roles Sample data for demonstration purposes only
- 39. 2 - Assign your channels and campaigns to See-Think-Do roles Assist Ratios A STEP BY STEP PROCESS ÷ = Last-Click Conversions Assist Metrics Assist Conversions Sample data for demonstration purposes only
- 40. A STEP BY STEP PROCESS 2 - Assign your channels and campaigns to See-Think-Do roles Display & Video: Lower Funnel 15.0 5.0 0.5 0.3 SEE Organic Search Direct 3.1 0.8 0.6 Assist Ratios SEE THINK DO Assist Conversions 9.0 Email Display & Video: Upper Funnel Display & Video: Mid-Funnel Paid Search: Generic 0.7 Referral Sample data for demonstration purposes only (may differ from previous/following screenshots) USE THE ASSISTED CONVERSION REPORT AND WORK ON YOUR CHANNEL GROUPING Brand Paid Search
- 41. A STEP BY STEP PROCESS 1 - measure the potential of attribution 2 - design a See-Think-Do channel grouping 3 - adapt your KPIs
- 42. A STEP BY STEP PROCESS 3 - Adapt your KPIs to SEE-THINK-DO performance Generic Search Click Assist interactions Referral Remarketing Brand Search Click Click Newsletter Signup Display Impression Microconversion Last interaction % new users SEE THINK DO Purchase Macroconversion Microconversion
- 43. A STEP BY STEP PROCESS 3 - Give monetary values to your engagement goals source: Dave McClure’s example conversion metrics
- 44. A STEP BY STEP PROCESS SEE, THINK and DO 3 - Adapt your KPIs to SEE-THINK-DO performance Acquisition ▼ Cost Analysis ▼ AdWords ▼ Display Targeting Choose Microconversion Acquisition ▼AdWords ▼ Display Reporting Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fall Collection (Int./Affinity) Fashion & Style - Chanel Fashion & Style - Michael Kors Fashion & Style - Steve Madden Fashion & Style - Dior Fashion & Style - Adidas Fashion & Style - Nike Fashion & Style - Reebok Goal 2: Store Locator Microconversions ANIMATION Sample data for demonstration purposes only SET MICRO-CONVERSIONS
- 45. A STEP BY STEP PROCESS 1 - measure the potential of attribution 2 - design a See-Think-Do channel grouping 3 - adapt your KPIs 4 - apply relevant models
- 46. USE MODEL COMPARISON TOOL on A STEP BY STEP PROCESS 4 - apply relevant models to new KPIs/Channels Conservative growth strategy your new KPIs and new CHANNELS (period = 10x avg path length) Aggressive growth strategy Finishing more existing conversion paths Starting more new conversion paths Converting more existing customers Converting more new customers Driving brand capitalization Driving brand awareness EXPERIMENT WITH A GRADUAL SHIFT AWAY FROM LAST CLICK
- 47. MOVE TOWARDS CUSTOM MODELS FOR MORE SOPHISTICATED ALLOCATION ANIMATION Sample data for demonstration purposes only
- 48. “THINK” CAMPAIGN SCORECARD COMBINING TRADITIONAL AND NEW METRICS Traditional Metrics Assist Conversions & Ratios Interest/Affinity Campaign 6.5 “Think” Average 3.2 $5K Investment 55k Clicks (5% CTR, 1.1M impressions) 19 Conversions (0.01% CVR, $500 CPA, 0.3 ROI) Page Depth & Microconversions 0.8 Store Locator views/session (+90% vs. site avg.) 150 Accounts created (10% CVR) 15 Potential future customers* *10% Avg. Conversion rate from microconversions to macroconversions over 6 month time period Assist Conversions Assist Ratio ● Assisted 5% of all transactions (125) ● Assisted 5% of all revenue ($11,125) ● 54 conversions with Think Custom model ● 2x higher likelihood of assisting (vs. ‘Think’ average) ● +40% higher avg. order value when assist Sample data for demonstration purposes only (vs. overall AOV)
- 49. DATA-DRIVEN MODELS ALGORITHM-POWERED MODELS IN GA PREMIUM 2% Search Email Likelihood of Purchase Display 3% Search Email Likelihood of Purchase Increased likelihood of purchase by 50% Premium
- 50. 25% growth in ROI x2 non-brand paid search performance x3 display performance +44% website sales
- 51. Want smarter attribution? Give it more data ! Demo of UA’s userID and Measurement protocol features
- 52. DOUBLECLICK INTEGRATIONS Premium import Impression data in your attribution reporting Count the DCM impressions in your attribution reporting Create remarketing lists in GAp and push them instantly in DBM
- 53. CROSS-DEVICE MEASUREMENT AD AD AD SITE APP SITE SITE APP User 1 User 2 User 3 USING THE USER-ID FEATURE DEMO
- 54. CROSS-DEVICE MEASUREMENT Creates profile based on User-ID instead of Client-ID USING THE USER-ID FEATURE
- 55. OFFLINE CONVERSION MEASUREMENT DEMO USING THE MEASUREMENT PROTOCOL SITE AD AD AD Online leads are qualified on the phone Sales close after in-person meetings Revenue generation happens in store attribute ?
- 56. EVEN BEYOND?
- 57. RETHINK YOUR MEDIA RETHINK YOUR METRICS RETHINK YOUR MODEL RETHINK DATA SCOPE JUST DO IT ! SO, WHAT TO REMEMBER NOW?
- 59. APPENDICES
- 60. USEFUL LINKS Framework and methodology related content Avinash Kaushik blog: http://www.kaushik.net/avinash/ Avinash Kaushik blog - framework post: http://www.kaushik.net/avinash/see-think-do-content-marketing-measurement-business- framework/ Dave McClure’s “start-up metrics for Pirates”: http://www.slideshare.net/dmc500hats/startup-metrics-for-pirates-long-version
- 61. OFFLINE CONVERSION MEASUREMENT USING THE MEASUREMENT PROTOCOL Code to extract GA’s CID of a user for later use in Measurement Protocol <?php function gaParseCookie() { if (isset($_COOKIE['_ga'])) { list($version,$domainDepth, $cid1, $cid2) = split('[.]', $_COOKIE ["_ga"],4); $contents = array('version' => $version, 'domainDepth' => $domainDepth, 'cid' => $cid1.'.'.$cid2); $cid = $contents['cid']; } else $cid = gaGenUUID(); return $cid; } ?> NON-GOOGLE CODE use at risk your own
- 62. OFFLINE CONVERSION MEASUREMENT USING THE MEASUREMENT PROTOCOL Example of Measurement Protocol HIT ( more info) = http://www.google-analytics.com/collect?v=1 &tid=UA-XXXXX-6 &t=pageview &dl=https%3A%2F%2F2.zoppoz.workers.dev%3A443%2Fhttp%2Fwww.yourwebsite.php &cs=WARSAW%20GAUC --- optional &cm=Jonathan --- optional &dt=testMP &cid=<?php echo gaParseCookie() ?> &uid=CRM_ID --- optional
- 63. LINKS TO CASE STUDIES Attribution success stories On the Beach: http://www.google.com/analytics/customers/pdfs/on-the-beach.pdf Sellpoints: http://www.google.com/analytics/customers/pdfs/sellpoints.pdf Amari: http://www.google.com/analytics/customers/pdfs/amari.pdf More case studies available at http://www.google.com/analytics/customers/