Managing Customers and Social Media ROI

Question 1

Social interactions take place on a one-to-one-to-many basis, meaning

Answer 1

the way in which the interaction is resolved can impact directly on brand sentiment.

Question 2

A customer service agent dealing directly with a customer on Facebook is engaging in a personal interaction with a public audience. It’s easy and logical for customers to want to

Answer 2

share socially about their service experience afterward, too.

Question 3

Key benefits of social CRM are that it enables a business to:

Answer 3

1. Deliver customer support in the online platforms customers use
2. Interact and engage with customers in real time
3. Resolve issues speedily by monitoring social media for complaints
4. Find and reward brand advocates and customers who help others
5. Get greater exposure in the places where audiences spend their time
6. Increase engagement and deepen relationships with customers

Question 4

Create a dedicated customer service channel on social media

Answer 4

 For example, create a Twitter account specifically for customer
service
 Add an option to your FB Messenger chatbot
 Use social listening to monitor brand buzz and to act on it
 Build an online brand community and invite your customers
to participate

Question 5

Present the social media intervention as a way to help

Answer 5

the customer service department meet its objectives

Question 6

 Customer service objectives

Answer 6

 Call volume, call time, calls handled, cost per call, resolution
rate, and hold times for a call center

 Time to resolution, response rate, time to response, and the
number of interactions before a resolution for other
organizations

Question 7

How to Measure ROI for SMM?

Answer 7

 What you’ve spent

 The sales that resulted (or business outcomes that
can be quantified)

Question 8

Barriers for ROI Measurement

Answer 8

 No industry standard
 Confusion between measurement and monitoring tools
 Social media silos
 Path to conversion with many touch points
 Attribution
 Your (disintegrated) software systems

Question 9

There is No Standard ROI Measurement

Answer 9

 The measurement changes from company to company, and
even from campaign to campaign
 Manager-level metrics vs. executive-level metrics
 Inertia to learn new metrics
 What is easy to track versus what is linked to sales
 Correlate the trackable social media metrics with measures
that executives care about: sales volume, revenue, and costs

Question 10

A tool can measure ROI only if

Answer 10

it actually connects to the system that

can track revenue.

Question 11

Monitoring tools are platforms for

Answer 11

scanning social media networks and

websites for related conversations

Question 12

Engagement tools are platforms for

Answer 12

publishing and replying to status

updates

Question 13

Measuring tools are

Answer 13

platforms for tracking the success of your social

media efforts that tie into conversions and provide reporting

Question 14

Social Media Silos

Answer 14

A starting point: to present social media alongside other marketing channels including
 Online advertising
 Offline advertising
 Search engine optimization
 Public relations
 Events
 Direct mail
 Email marketing

Question 15

The roles played by different channels

Answer 15

 Initiate
 Assist
 Complete

Question 16

Difference between ROI and Attribution

Answer 16

 ROI is a measure of how much you earn based on how much
you spend
 A dollars-and-cents measure of the return per unit investment

 Attribution assigns value to the channels/venues/campaigns
that drove an outcome, whether it is revenue-related or not.
 Often involves working backwards from an event and figure out
what channel(s) get the credit for that event, and how much

Question 17

Software packages marketers use are

Answer 17

separate and do not talk
to each other
 Create measures that take a bird’s-eye view
 Can’t pass data back and forth across systems easily
 Integration is needed to tie actions at individual customer
level

Question 18

To understand what the path to conversion might look like

Answer 18

 List all the marketing communication vehicles a prospect could
touch during the path to conversion into a customer

 List all the marketing communication vehicles a customer could
take throughout the customer life cycle

Question 19

How do we track full campaign history for prospects and

customers?

Answer 19

 Identify possible touch points in the campaign
 Integrate the campaign history into the CRM
 The best way to do that is through a trackable URL
 Track URLs that drive to your own web properties and determine whether
those URLs will enable you to understand users’ downstream activities

Question 20

Collecting campaign data

Answer 20

UTM in Google Analytics: Urchin Traffic Monitor (UTM) fields are added to
the end of the URL, UTM parameters are recognized within Google Analytics,
and passed through the reporting.

Question 21

 5 UTM parameters you can add:

Answer 21

 Source: identify the advertiser, site, publication, etc.
 Medium: the advertising or marketing medium, for example, banner, email
etc.
 Campaign: the individual campaign name, slogan, promo code, etc.
 Term: identify paid search keywords
 Content: used to differentiate similar content, or links within the same ad

Question 22

Social Reports feature in Google Analytics

Answer 22

 When SM is a top-of-funnel contributor, it would be identified
as a “social assist” in this reporting

Question 23

Tracking traffic you sent to third-party websites

Answer 23

 Intercept them before they get there and put your tracking
parameters in place
 Build a custom URL shortener that intercepts this process and
drops in the tracking code before they go to the third-party site.
 For example, append Google’s custom URL parameters before
the link is shortened

Question 24

Rule Attribution Model

Answer 24

 Adopted by most Web Analytics Providers – IBM, Adobe, and Google

first touch

last touch

even

time decay

weighted model

algorithmic

Question 25

Credit of channels for the path:

Answer 25

display → referral → email → search → purchase

Question 26

Rule-Based Attribution Model

 Advantages:

Answer 26

 Easy to apply;

 Can be powerful when combined with big data

Question 27

Rule-Based Attribution Model

 Disadvantages:

Answer 27

 Disregard paths that do not lead to conversion

 Inconsistent and inconclusive

Question 28

Using your complete campaign history data

Answer 28

• Build a model considering all conversion and non-conversion paths
  and estimate the contribution of each touch point
• Based on the parameter estimate optimize the touch point deployment

Question 29

Li, Xie and Zheng 2019, MIS Quarterly

Answer 29

 Customers are targeted by multiple firms.
 Their interactions with competitors will affect your ad effectiveness.
 Single-firm model versus competitive model
 A purchase funnel view
 Customers narrow consideration set.
 Channels play different roles in the customer journey.
 Loyalty affects repeat purchase.

Question 30

Panel data from a leading Internet analytics company

Answer 30

 19,003 purchase related search sessions in flight ticket booking
 Browsing behavior within 7-day window
 4,897 sessions ended with a flight ticket purchase

Question 31

4 channels

Answer 31

 Search engine, email, display ads/referrals, direct

Question 32

10 competing (sets of) websites

Answer 32

 5 major online agents (cheaptickets.com, expedia.com,
priceline.com, orbitz.com, travelocity.com) + small agents + 3
top airline companies’ direct websites + other airlines

Question 33

Data Pattern

 Limited search

Answer 33

 Limited search: 34.3% of consumers only visited one website.

 Being the entry website is important: 59.6% of all transactions were made
on the first website visited during a session.

Question 34

Multichannel Advertising

Answer 34

Awareness :
Which websites do I know?

Alternative evaluation stage:
Which websites should I visit?
–Entry Website–>Visit Decision on Remaining
Websites

Transaction stage:
Which website should I buy
from?

Question 35

Two sources of heterogeneity

Answer 35

 Preference for different websites

 Consideration set

Question 36

The Model – Visit Stage

Answer 36

 Choice of entry site
 Out of all websites in the awareness set
 Model: Multinomial Logit with random coefficients

• Whether to search other websites in the awareness set or not
 Model: s-1 independent binary Logit with random coefficients

 Covariates (for both sub-decisions)
 Number of channel specific clicks in the previous purchase session
 Advertising channels: search engine, email, display ads/referrals
 State dependence
 Lag search
 Lag purchase
 Past experience
 Cumulative pages browsed on the website
 Cumulative money spent on purchases through the website

Question 37

The Model – Purchase Stage

Answer 37

 Choice of conversion website
 Out of all searched websites
 Model: Conditional Logit with random coefficients
 Covariates
 Number of advertising clicks in the current purchase session
 Advertising channels: search engine, email, display ads, referrals
 State dependence
 Past Experience
 Cumulative pages browsed on the website
 Cumulative money spent on purchases through the website
 Position in search sequence (purchase stage only): first / last

Question 38

The information stock (interactions) from all channels have

Answer 38

significant positive effects at both the visit stage and the purchase stage.

 Direct > search > email > display/referral

Question 39

Return metrics for social media: if not

directly lead to sales

Answer 39

Return on
impressions
model

Return on
social media
impact model

Return on
target influence
model

Return on
earned media
model

Question 40

Using these alternative return metrics

Answer 40

Cost-benefit analysis