C O N F I D E N T I A L Marketing Analytics Get to Grips with Attribution Modeling Jonathan Isernhagen EyeForTravel Social Media & Mobile Strategies for Travel 3/18/2014
C O N F I D E N T I A L
Marketing Analytics
Get to Grips with Attribution Modeling
Jonathan Isernhagen
EyeForTravel
Social Media & Mobile Strategies for Travel
3/18/2014
C O N F I D E N T I A L
Get to Grips with Attribution Modelling Pinpoint What Really Influences Customer Behavior
1) What useful tools are available to help you accurately attribute
influence? How can you measure sentiment?
2) How do you communicate ROI for mobile and social back to upper
management?
3) Are you interpreting conversion cross platform accurately?
4) Is visual social media harder to measure than other forms?
C O N F I D E N T I A L
Effective Influencers have…
C O N F I D E N T I A L
• Appinions
• eCaim
• Empire Avenue
• Klout
• Kred
• mPACT
• PeerIndex
• PROskore
• Radian6
• Traackr
• TweetLevel
• TweetReach
• Twitalyzer
• TwitterGrade
Influence measurement vendors
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
Brian Solis, http://www.briansolis.com/2012/03/report-the-rise-of-digital-influence/
C O N F I D E N T I A L
The KPIs behind the KPIs
Influence KPIs:
• Brand Lift/Awareness: aided/unaided recognition of brand.
• Sentiment Shift: opinion movement from negative to positive
• Brand Resonance: how often is your brand discussed?
• Reach-through advocacy and Word-of-Mouth: virality
• Leads/Referrals: leads driven by promotions
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
Business goals of digital influence
• Boost brand awareness, leading to sales
• Attenuate brand scandals (e.g. “United Breaks Guitars”) by
pre-emptively co-opting some influencers, saving sales.
C O N F I D E N T I A L
Get to Grips with Attribution Modelling Pinpoint What Really Influences Customer Behavior
1) What useful tools are available to help you accurately attribute
influence? How can you measure sentiment?
2) How do you communicate ROI for mobile and social back to upper
management?
3) Are you interpreting conversion cross platform accurately?
4) Is visual social media harder to measure than other forms?
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
ROI calculation/communication
Channel* Desktop Tablet Mobile
Brand 20%
Brand SEM 62% 51% 38%
Display -5% -12% -7%
Display - Retargeting 26% 25% 29%
Email 250%
Meta search 18% 22% 10%
Non-brand SEM -30% -18% -40%
SEO 500% 520% 390%
Social media -5% -15% 15%
- 7 -
*of impression/click, not necessarily of consumer conversion
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Positive Return on Incremental Channel Investment
- 8 -
ROI = (VCM – Spend)
Spend
Abbreviation Term Definition
ROI Return on Investment Indicator of investment profitability.
Positive = good.
VCM Variable Contribution Margin The amount of profit driven by a
given transaction.
Spend Channel Spend The amount spent driving traffic to
the site during the period in question
Calculated over a specified time period of investment and return.
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
ROI = (VCM – Spend)
Spend
Spend
- 9 -
Which spend do you include?
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Marketing ≈ Farming
1) Seed = impressions
2) Bookings = fruit
- 10 -
But….
marketers plant
& harvest constantly.
C O N F I D E N T I A L
spend spend spend spend
How changes in spend can mess up ROI
- 11 -
$
time
profit
impact impact
spend
impact
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
Jan Feb Mar
C O N F I D E N T I A L
Variable Contribution Margin (“VCM”)
- 12 -
ROI = (VCM – Spend)
Spend
1) What is the value of each transaction?
2) Which transactions go to which channels?
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
VCM
Transaction revenue minus variable non-marketing expenses:
• Revenue: –Supplier Commissions,
–GDS incentives,
–Overrides (lumpy: average/booking must be assumed)
–Media (not transaction-driven, but has to be modeled in somewhere)
–Attached bookings / Lifetime value: try to gauge value without double-counting
• Expenses: –Website hosting/capacity costs
–Data processing expenses
–Other expenses which vary by transaction or site activity volume
- 13 - [email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Which marketing treatment(s) triggered the purchase?
"Half the money I spend
on advertising is wasted;
the trouble is,
I don't know which half.“
-John Wanamaker
Father of Modern Advertising
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
“Hard to Know” ≠ “Random”
We may not find the correct answer, but the answer does exist.
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
• Sends
• Opens
Collecting the necessary data
Analysis
space
• Clicks
• Visitors
• Transactions
• VCM
• Media impressions
Site metrics tool
Back office system
• Channel-specific
phone #s
Email service provider
Call Center IVR
• Impressions
• Clicks
• Spend
Display ad server
e.g. SAS, Revolution Analytics,
SPSS, Teradata Warehouse Miner
SEM bid management tool
• Impressions
• Clicks
• Spend
• GRPs
• Spend
Television plan
• Impressions
• Clicks
• Spend
Meta search feeds
Spend
Manual spend entry table
C O N F I D E N T I A L
Connecting the necessary data
Transactions Profiles Customers
Sessions
∞ 1 ∞ 1 Clicks:Transactions Clicks
∞ 1 1 ∞
∞ 1 ∞ 1
Calls
Sends
Opens ∞
1
∞
1
IVR
GRPs
TV plan
Impressions
Ad server
Impressions
∞ 1
∞ 1
∞ 1
Back office Site monitoring tool CRM system
1 ∞
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Get to Grips with Attribution Modelling Pinpoint What Really Influences Customer Behavior
1) What useful tools are available to help you accurately attribute
influence? How can you measure sentiment?
2) How do you communicate ROI for mobile and social back to upper
management?
3) Are you interpreting conversion cross platform accurately?
4) Is visual social media harder to measure than other forms?
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Profile deduplication
Profiles Customers ∞ 1
CRM system
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
1) App registration information tied to desktop profile
2) Logged-in customer information on multiple devices
3) E-mails sent to same address opened on multiple devices
4) Third-party services with network visibility
C O N F I D E N T I A L
Get to Grips with Attribution Modelling Pinpoint What Really Influences Customer Behavior
1) What useful tools are available to help you accurately attribute
influence? How can you measure sentiment?
2) How do you communicate ROI for mobile and social back to upper
management?
3) Are you interpreting conversion cross platform accurately?
4) Is visual social media harder to measure than other forms?
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Social attribution requires impressions
Transactions Profiles Customers
Sessions
∞ 1 ∞ 1 Clicks:Transactions Clicks
∞ 1 1 ∞
∞ 1 ∞ 1
Calls
Sends
Opens ∞
1
∞
1
IVR
GRPs
TV plan
Impressions
Ad server
Impressions
∞ 1
∞ 1
∞ 1
Back office Site monitoring tool CRM system
1 ∞
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
SocialCode clicks vs. impressions example
This example overcredits impressions (excludes other channels)
but it gives some idea of the sub-surface portion of the
impressions iceberg.
C O N F I D E N T I A L
Facebook impression/click visitor id tracking
Visitor action Volume Trackable Useful
Ad link click Small Yes No
Sponsored news feed click Small Yes No
Organic click Small Yes No
Ad impression Large Yes Yes
Sponsored news impression Large No Yes
Organic impression Large No Yes
With one exception, what’s useful is not trackable,
and what’s trackable is not useful.
C O N F I D E N T I A L
Forrester 2012Q2 survey of attribution vendors
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Algorithmic attribution per Visual IQ: Top Down (MMO) & Bottom Up (Fractional Attribution)
25
SUMMARY LEVEL DATA USER LEVEL DATA
CHANNELS Offline + digital cross channel Digital media channels
ROLE Strategic: Optimize spend across channels
Tactical: Generates granular
media recommendations
OUTPUT Cross channel insight Full fractional attribution
CAPTURES Seasonality and external factors
Interplay between digital touch
point and channels
PREDICTABLE
GRANULARITY Conversions at aggregate level Propensity to convert at user level
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
“Given all the history we know, how likely is this shopper to convert soon?”
Datasong answers with a 2-stage model:
-Are Brand shoppers (BSEM, BSEO, DTI) likely to book? Yes
-Did your last TV spot cause their brand loyalty? Maybe
Algorithmic attribution per DataSong: Survival Modeling
Model 2 Accuracy 81%
Model 1 Accuracy 68%
0
0.5
1
0 0.5 1
NonConverters Converters
0
0.5
1
0 0.5 1
NonConverters Converters
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Algorithmic attribution per DataSong: Survival Modeling
1) Axes show the time since last channel exposure
2) Dots represent converters vs. nonConverters
3) Orange line represents the relationship between 2
variables, e.g.: a) time since last email and
b) time since last affiliate visit
4) Responders: everyone above the line. We count the
folks above the line and see what our accuracy is
5) Attribution: once satisfied with a model(s), we’d: a) take a given order,
b) see the time since last email and affiliate and based on
the timing, and;
c) where we are on the orange line, so;
d) we have a means to allocate which was more causal.
0
0.5
1
0 0.2 0.4 0.6 0.8 1
NonConverters Converters
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Survival Modeling: DIY
Recommended by one of our
PhD statisticians.
“SAS usage is not necessary.”
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L - 29 -
Algorithmic attribution per Google: Interaction Method (“Shapley value”)
Brand
SEM Email
2%
Conversion
Brand
SEM Email
3%
Conversion
Every combination (“coalition”) of clicks is tested.
“How important is each player to the overall cooperation?”
http://en.wikipedia.org/wiki/Shapley_value
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
Display
C O N F I D E N T I A L
Running the process
1) For each site visitor a) Assemble visit history
b) Create variables to represent: i. Channel impressions
ii. Channel clicks
iii.Past purchases
2) Regress or use machine learning algorithm a) Ascertain which channel touches predict booking
b) Give VCM credit to causal channels
3) Calculate ROI a) Use each channel’s VCM and spend
b) Where ROI is positive, spend up
c) Where ROI is negative, cut spend or change tactics
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Take-away lessons
- 31 -
1) Tie your social media investments to the bottom line. 1) Do the best attribution you have resources for
2) Be mindful of what data you’re leaving out
3) Click basis will undervalue social contribution
2) Experiment to discover each influencer’s value
3) If your data aren’t yet in shape to support attribution & other ad hoc analyses, get that project underway.
[email protected] @jon_isernhagen www.linkedin.com/in/isernhagen
C O N F I D E N T I A L
Top Social Media Sites
Site Focus/Description UVs/Month
Facebook Sharing blog postings, photos/videos, apps w friends 900M
Twitter 140-character micro-blogging, RSS, updates 310M
LinkedIn Business and professional networking 250M
Pinterest Online pinboard for organizing and sharing images 150M
Google Plus+ Social layer atop Google properties, authorship tool 120M
Tumblr Microblogging platform and social networking 110M
Instagram Square-shaped, filtered photo and video sharing 85M
VK Russian speakers’ Facebook 90M
Flickr Photo sharing, hosting, commenting, networking 65M
MySpace General: Photos, Videos, Blogs, Apps. 40M
Tagged Social discovery website for game/gift sharing 38M
Ask.fm Anonymous question/response site 37M
Meetup Offline meeting planning for various activities 35M
MeetMe Social networking/introduction site with games 10.5M
ClassMates Seek/connect w school, work, military friends 10M
C O N F I D E N T I A L
Attribution options - broad view
Provider Method Cost
Timing Comments 1x monthly
Adobe Attribution model Varies - 4-10 wks Consulting arm willing to address attribution questions
as one-off modeling projects.
Adometry Econometric
model Low High 12 wks
Forrester-acclaimed industry co-leader.
Highly display-focused.
Datasong
Attribution model
w/
Customer
scoring
Med
Med
6 wks
Comprehensive model which pulls in both online and
offline spend impacts.
Med Per-customer demand curve to predict the expected
return on every campaign
Analytics Interactions
model - - now
Has delivered digital-only channel interactions model
to beta clients.
Gauge LLC Simple
regression High - 10 wks
General consulting analysts, not on the efficient
frontier for this deliverable.
Kenshoo Unknown
algorithm - Med 4 wks
Primarily targeted towards SEM spend, but can be
expanded to comprise other digital channels
Visual IQ Econometric and
Attribution
models
High - 4-8 wks Econometric model. Offers cross-channel allocation
guidance, including offline spend.
Med-
High - 3-6 wks
Attribution model. Offers within-channel campaign-
level spend guidance. No offline.