Download to read offline
![1. First touch attribution
2. Lead creation touch attribution
3. Last touch attribution (Opportunity creation touch)
4. Last non-direct touch attribution
5. Last [Insert Marketing Channel] Touch Attribution
6. Linear Attribution
7. The time decay conversion
8. U-Shaped (Position-Based) Attribution
9. W-Shaped Attribution
10. Full-Path (Z-Shaped) Attribution
Marketing attribution models](https://image.slidesharecdn.com/firstpartycookies-forblog-210217104350/75/Understanding-first-party-cookies-and-attribution-models-29-2048.jpg)
![5. Last [Insert Marketing Channel] Touch
Attribution
• This is the last Adwords touch model, Facebook or
Twitter touch model.
• This is the last touch before whatever conversion you
configured the analytics to measure such as lead
conversion, or opportunity conversion…
• While each attribution comes standard with their
channel as each insights, either on FB or Adwords, uses
a Last Adwords Touch Model, it’s biased to their own
channel.
• For example, if a visitors clicks on a Facebook ad on
Monday and then a Google ad the next day and then
converts, both channels model credit 100% of the
conversion credit.](https://image.slidesharecdn.com/firstpartycookies-forblog-210217104350/75/Understanding-first-party-cookies-and-attribution-models-34-2048.jpg)
Uploaded byVbout.com
Understanding first-party cookies and attribution models
The document discusses the shift from third-party cookies to first-party cookies due to increasing data privacy regulations and consumer demands for data transparency. It outlines the importance of identity resolution in marketing, contrasting deterministic and probabilistic methods, and presents several marketing attribution models to analyze customer interactions and optimize marketing strategies. The need for companies to adapt their advertising and data practices in a post-cookie world is emphasized, along with strategies to encourage consumer opt-ins for first-party data usage.
More Related Content
Similar to Understanding first-party cookies and attribution models
![[Webinar] How the Cookie Crumbled: Preparing for a Time without Third-Party C...](https://cdn.slidesharecdn.com/ss_thumbnails/nahowthecookiecrumbledwebinar-2-210312165823-thumbnail.jpg?width=640&height=640&fit=bounds)
Understanding first-party cookies and attribution models
- 1. Understanding first-party cookies andattribution models C O M M U N I T Y
- 2. 15K member. 6years C O M M U N I T Y
- 3. • 5:00-5:20: Attendeesintros • 5:20 – 5:35: A guide to the post-cookie world • 5:35-5:55: Open mic • 5:55-6:00: QA Agenda
- 4.
- 5. Understanding first-party cookies and attribution models A guideto the post- cookie world
- 6. Intro • With thedeath of third-party cookies and the evolvement of data privacy and legislations such as GDPR and CCPA, companies need to be aware of these in order to future-proof their technology. • First-party identity resolution or the ability to stitch all customer data into accurate profiles is key to improve marketing insights, optimize ad spend and deliver flawless customer experience.
- 7. Difference between first-partyand third-party cookies • For first-party cookies, the cookies are created and stored by the website the user visits directly. They enable website owners to collect analytics data, remember language settings, and perform functions that help create a good user experience. • For first part-cookies, engagement is happening with the same website, app or endpoint. • Third-party cookies are created by any domain other than the website. It can be cross-site tracking, retargeting (ads) and ad-serving (display ads).
- 8. Source: us.epsilon.com Difference betweenfirst-party and third-party cookies
- 9. Why third-party cookiesare dying • As the emphasis on user privacy is growing, third-party cookies are becoming a general threat to targeted advertising and marketing. • A growing number of consumers are not comfortable with the way their data is shared. They want transparency, choice and control over how their data is used. • Safari and Mozilla Firefox no longer support third-party cookies. • These cookies will be also blocked in Chrome by 2022.
- 10. Third-party cookies deathconcerns • Advertisers are confused about the future of advertising along with the death of third-party cookie. • Ad tech companies like Facebook, Google and Amazon collaborate on a collective cookie replacement. • In light of these, companies started to recognize the importance of having high quality identity graph of their own customers.
- 11. What’s next? Without Chrome-basedthird-party cookies data, you’ll still be able to make use of Google Ads targeting, which will be powered by Chrome’s first-party cookies and the Privacy Sandbox tools.
- 12. What are yourcompany’s/main client’s tactics for addressing identity resolution in the future?
- 13. • Deterministic: IDresolution uses what you know to be true. It merges identifiers like phone numbers, emails, device IDs and user IDs that you have collected from the user. • It’s a high-confidence approach that users first-party data where you definitely know with certainty the user’s behavior and data points. • Probabilistic: ID resolution uses what you assume to be true. Using predictive algorithms and/or third-party databases, it attempts to merge different identifiers to piece together a single view of the customer. Deterministic Vs Probabilistic Methodology
- 14. Source: Segment Deterministic VsProbabilistic
- 15. • Intelligent trackingprevention is a featured offered with Safari and iOS 11 by default. • It changes the way Apple handles first-party cookies unlike most other browsers. • Its newest version, ITP 2.0, detects cross-site tracking, which makes it impossible to use them in third-party context for tracking or analytics purposes. This is different from their previous versions (1.0 and 1.1) where first-party cookies could be used in a third-party context. Apple’s Intelligent Tracking Prevention (ITP) and cookies
- 16. Source: Segment Evolvement ofcookies and data privacy legislations
- 17. How to driveprospects to opt in and make use of first- party data?
- 18. 1. Browser pushnotification • SEO, PPC or organic social media efforts driving prospects to your website where they accept to receive notifications from you regularly. • Accepting cookies and getting daily notifications via browser push is a successful practice of using first-party cookies.
- 19. • People canvisit a landing page or website via an ad or other marketing initiatives where they are asked to fill out a form or subscription bar to receive blog newsletters. They can be also redirected to an article where a popup shows. • Once they agree to the company’s terms of service and privacy policy or confirm that they are humans, it means that their info has been successfully captured via first- party cookie. 2. Landing page, newsletter forms and popups
- 20. The 3 usecases an effective identity resolution helps solve • Analyze end-to-end customer journeys • Optimize your marketing efficiency • Deliver consistent experiences
- 21. • Some prospectsand visitors might be anonymous as they haven’t provided their contact info to the company yet. • Therefore, it’s critical to tie the unknown user activity to a user profile once has made herself known. • Without an accurate system that ties unknown and known users, end-to-end customer journey mapping becomes impossible because the connection between anonymous and post-sale activity is broken. • Therefore, with the right identity resolution, marketers, analysts and product owners can invent new features, optimize site experience and create powerful retention campaigns. 1. Analyze end-to-end customer journeys
- 22. Example of marryingan anonymous IP with their known user profile
- 23. • A properidentity resolution that stitches together accurate customer profiles in real-time, enables marketers to create high-performing campaigns that save valuable marketing budget. • According to Gartner, the average brand spends 14% of its entire marketing budget on personalization tactics. • Marketers can initiate powerful omni-channel campaigns when there’s one centralized repository of customer data from all sources. • Whether it’s advertising, email or support centers, marketers can deliver the right message, to the right person at the right time. 2. Optimize your marketing efficiency
- 24. Tying a newuser profile of the same customer to their existing known profile
- 25. • Customers nolonger use a single touchpoint; they rather engage on mobile apps, mobile browsers, desktop and more. • A proper identity resolution allows marketers to create a consistent experience across all their touchpoints. For example, if a user purchases a product on their phone, they shouldn’t see an ad while browsing on their desktop afterwards. • Identity resolution enables companies to unify all their user across devices and channels into one single view of their customer. 3. Deliver consistent experiences
- 26. Unifying data acrossmultiple touchpoints into a single view
- 27. Achieving a singleview of the customer • Company teams can tap into customer data platform (CDP) to create a personalized customer experience. • This is done through identity resolution, which unify all customer touchpoints across channels, devices, and online and offline activity. • Stitching all data leads to create comprehensive profiles that grow with customers over time.
- 28. • The importanceof marketing attribution is to study how customer touchpoints drive conversions. • It measures each channel effectiveness and assigns credits to the touchpoints that lead the customer to convert. • This allows marketers to better map their customer journey and optimize their marketing campaigns. Marketing attribution
- 29. 1. First touchattribution 2. Lead creation touch attribution 3. Last touch attribution (Opportunity creation touch) 4. Last non-direct touch attribution 5. Last [Insert Marketing Channel] Touch Attribution 6. Linear Attribution 7. The time decay conversion 8. U-Shaped (Position-Based) Attribution 9. W-Shaped Attribution 10. Full-Path (Z-Shaped) Attribution Marketing attribution models
- 30. 1. First TouchAttribution • This model gives credit to all the marketing efforts that pushed the prospect to interact with the first page. • It gives emphasis on the top of the funnel where awareness is created. • Its drawback is the inaccuracy between the first touch and conversion. For example, as the cookie expires within a 30 to 90-day period, this model is giving attribution to the first touch within this period and NOT the true first touch where conversion might take greater than 90 days. (i.e: If you are using conversion tracking using Google Analytics).
- 31. • This modelhelps you determine the marketing channels that drive lead conversions. • However as there’s a lot more than just a lead signup or creation, this remains a small part in the customer journey so it’s unrealistic to give 100% credit to lead creation only. • Also this model may often be confused with the first touch model because marketing automation platforms consider that the first session starts from whenever the lead is created. 2. Lead Creating Touch Attribution
- 32. • It’s thesimplest model for attribution systems. • It assigns credit where a sale opportunity is created and where the end of the marketing funnel is. • It also has the smallest time window for an error to occur as the period is very short, unlike the first touch model. 3. Last Touch Attribution (Opportunity Creation Touch)
- 33. • This issomehow more useful than the simple last touch model, because it doesn’t take into account the “Direct Traffic” which may not come from manually typing the URL but also from any traffic that doesn’t have a referral source. For instance, any social posts, social ads, or emails that don’t have a proper UTM tracking is considered direct traffic. • As a result, this model helps you avoid the troubles of “Direct Channel” data which is often deceiving. 4. Last Non-Direct Touch Attribution
- 34. 5. Last [InsertMarketing Channel] Touch Attribution • This is the last Adwords touch model, Facebook or Twitter touch model. • This is the last touch before whatever conversion you configured the analytics to measure such as lead conversion, or opportunity conversion… • While each attribution comes standard with their channel as each insights, either on FB or Adwords, uses a Last Adwords Touch Model, it’s biased to their own channel. • For example, if a visitors clicks on a Facebook ad on Monday and then a Google ad the next day and then converts, both channels model credit 100% of the conversion credit.
- 35. • This methodgives credit to every single touchpoint in the buyer journey. • The good thing about this method is that it’s a multi- touch model. • Its shortcoming is that it doesn't take into account the varying impact of each touchpoint and may assign the highest credit to a channel that impacted the least in driving conversions. • For example, a prospect initially clicks on your landing page via your FB ad, stays on the page for an hour. 1 day later, they visit your site via direct search and then convert. Your landing page will get 5% of the credit while the remaining 95% goes to direct search. 6. Linear Attribution
- 36. • This modelassigns attribution to the channels or touchpoints closest to conversion. It makes the assumption that the closer to conversion, the more influence it has. • The main problem with this model is that it ignores top of the funnel as it considers it away from conversion. 7. Time Decay Attribution
- 37. • This modelfocuses on lead generation. • It’s a multi-touch model that tracks every single touchpoint but emphasizes the importance of 2 key touchpoints: the anonymous first touch that brought the visitor to the door and the lead conversion touch. • Each of the these touches are given 40% of the credit and the remaining 20% is split equally across the other touchpoints. • The problem is that it ignores marketing efforts beyond the conversion stage. 8. U-Shaped Attribution (Position-Based)
- 38. • This modeltakes takes the U-Shaped model to the lead opportunity stage, which is considered the end of the funnel for many organizations. • The W-Shaped gives emphasis to 3 touchpoints: First touch, lead conversion and opportunity creation touch. • Each of these 3 receives 30% while the remaining 10% is distributed equally among the remaining touchpoints. 9. W-Shaped Attribution
- 39. • Taking itone step further, the Full-Path model or Z-Shaped takes the attribution to include the step beyond the opportunity stage: the customer close. • In this model, each of the four touchpoints gets 22.5% of the credit and the last 10% is split equally among the remaining touchpoints. • This model is fair and accurate for organizations that do marketing to existing sales opportunities. • Adopting this method requires a close coordination between marketing and sales. 10. Full-Path Attribution (Z-Shaped)
- 40. Source: Ruler Marketing attributionsoftware and comparison
- 41.
Editor's Notes
- #12 Privacy Sandbox represents an alternative pathway that Google is providing for the ad industry to take, relying on anonymized signals (that are not cookies) within a person’s Chrome browser to profit from that user’s browsing habits. “The most significant item in the Privacy Sandbox is Google’s proposal to move all user data into the browser where it will be stored and processed,” said Amit Kotecha, marketing director at data management platform provider Permutive. “This means that data stays on the user’s device and is privacy compliant. This is now table stakes and the gold standard for privacy.”
- #23 For example, let’s say Jane Doe visits the e-commerce site of a hypothetical shoe brand called SegKicks. Like many internet shoppers, she doesn’t register for an account, but clicks on a few different types of shoes—ShoeA, ShoeB and ShoeC—but doesn’t add them to her cart. Because Jane hasn’t registered for an account yet, a new user profile is created with an anonymous ID. Let’s say Jane then decides to add ShoeD to her cart in the same session. At checkout, she purchases the shoes and creates a new account with her email. Now that she’s created an account, her email address and newly assigned user ID are tied with the previously anonymous user profile. Thus, instead of having two different user profiles —“logged-out Jane” and “logged-in Jane”— there is only one in the system. From here on out, Jane will have a single user profile that can grow with her as she continues to interact with the business, no matter what device or channel she’s using. As Jane’s customer profile grows over time with additional interactions and purchases, SegKicks is able to power journey analytics and lifecycle campaigns
- #25 Returning to the Jane Doe example, let’s say she downloads SegKicks’ companion app, SegRuns, on her phone to track her runs. This time, a new user profile with an anonymous ID and iOS device ID is created when she first opens the app. In order to start tracking her runs, Jane registers for an account with her email address. Because the email address is associated with a SegKicks account, her existing user profile is updated to include this new anonymous ID and iOS device ID from the SegRuns app. Having this real-time, cross-channel single view of the customer makes SegKicks marketing campaigns a lot more effective. Now, Jane’s e-commerce events from SegKicks and activity from SegRuns are linked into a single customer profile accessible across channels and devices.
- #27 Now, let’s say Jane wants to explore her monthly running activity on SegRuns. Instead of using her phone, she decides to download the app on her Galaxy tablet. Like before, when she first opens the app, the system would register a new anonymous user with an associated Android ID. Once she logs in to the app with her email address, her visit would then be authenticated to her existing User ID and email address. Like in our previous example, identity resolution would merge the identities, including all traits and behavior, together into a single customer profile. Regardless of device, Jane is the same person. With identity resolution, companies can manage and configure cross-device identity management in real time to deliver a consistent experience, wherever the customer is. They’ll know not to advertise the SegRuns app to Jane on any device or channel — she’s already downloaded the app. They’ll also know when to send her targeted ads for runnings shoes; as soon she’s logged 50 miles, SegKicks can automatically suggest she repurchase shoes via email, advertising, or even in the SegRuns app itself.
- #41 https://www.ruleranalytics.com/blog/analytics/marketing-attribution-software/