Social Media & Web Analytics | Summary
Block 4 – 2022/2023
WEEK 1 – THE SOCIAL MEDIA LANDSCAPE
Social media informs purchasing
- We use Social network Sites (SNS) to research products we want to buy
- We use SNS to discover brands / products we otherwise wouldn’t be aware of
- We use SNS to buy products
User Generated Content (UGC) = content generated or created by an internet user who is a
consumer of this information or content. They’re contributing back to the platform.
Social Media = the online platform that hosts this User-Generated Content.
- With this broad definition, we include many platforms that we traditionally don’t view as
Social Media (e.g. NYTimes, Uber, Rotten Tomatoes, Amazon, StackOverflow and WordPress)
Reasons why Social Media has taken off:
- Digital Social Networks: for a structured information flow around us
- Machine Intelligence: the ease and speed of recommendations they give
- Smartphones: we’re always on, and they send us notifications which we like
Aral’s Hype Machine: the previous 3 factors are layered on top of each other, to facilitate Social
Media growth.
- Because of the Digital Social Networks, we’re digitally connected to each other that’s
observable via social graphs.
- Machine Intelligence takes this information and guides consumers through content they may
want to consume (friends they may want to follow). Which is now doable because:
o Richness of the data: the network is now observable
o Advances in computational power: Smartphones are fast enough to do this
- Which is then used to give us notifications / signals via our Smartphone 24/7. Our brain likes
these signals / reactions we get (likes gives dopamine).
o We also use our smartphone to input data, by writing on Social Media, and thereby
continuously feed need information into the network, which is again used by
machine intelligence to create better recommendations.
Dark side of Social Media:
- Mental health effects: social media is detrimental to our health
- Echo chambers due to self-selection into content and recommender systems
o Different (groups of) individuals do self-selection into what we read. And
recommender system will keep on sending this information tailored to our needs,
that in the end, we do not get a clear broad view of what’s happening.
o This also happens in politics: political polarization (where you as a democrat only see
posts putting democrats in a positive light).
- Privacy concerns: we’re not aware how many data is collected on us online.
Why do people contribute to social media? Why do they want an audience?
- Intrinsic utility: inherent satisfaction
o Users receive direct utility from posting content
o An exogeneous increase in followers, is expected to lead to more posting.
, o Because we’re happy that we’re broadcasting to more people, which is an incentive
to post more.
- Image-related utility: status-seeking / prestige motivation
o We’re motivated by perceptions of others
o An exogeneous increase in followers, is expected to lead to no change or even
decrease in posting behavior
o Because we post to attract followers, and once we have them (a minimum number of
social bonds), it’s enough.
STUDY – TOUBIA & STEPHAN (2015) | INTRINSIC AND IMAGE-RELATED UTILITY
- A field experiment on 2500 twitter accounts
- They used 1335 active users, and gradually added 100 new followers to 100 of these
accounts.
- Results:
o No main effect on positing activity: there was no significant difference between those
100 accounts (treatment) and the other 1235 (control) in posting rate.
o A differential effect based on the number of followers started with
▪ Quintile 2 (out of 5): relatively low # followers leads to an increase in posting
rate (= intrinsic-utility effect)
▪ Quintile 4: relatively high # followers leads to a decrease in posting rate (=
image-related utility)
- Managerial implications:
o Brand advocates aren’t always those with large follower counts
▪ Image-related utility concerns start to come into play
▪ Micro / nano influencers may work better in that case
o As a social media platform matures (getting more and more followers), we see more
firm generated content
▪ Consumers stop broadcasting because of image-related utility
▪ Firms no longer use it as a social listening platform, but an advertising
platform.
Lurker = a user of a social media site who doesn’t actively participate / contribute, but they do
consumer information
- 90-9-1 rule: 90% of users are lurkers, 9% contribute from time to time, 1% account for the
most contributions
- But these numbers aren’t always correct, they change a lot.
- But the exact distribution doesn’t matter, as long as we do take into account that lurking
matters and it has a huge impact.
The marketing value of lurkers (Chen, 2019)
- Lurkers aren’t low value users, they still engage and make decisions based on what they see
online
- There’s a distinction:
o Passive lurkers = absorb content and don’t spread information
▪ They generate value from their own decisions
o Active lurkers (diffusers) = transmit information to others (after reading)
▪ They generate value from
• Their own decisions
• Diffusion: they share information with others offline
STUDY – CHEN (2019) | SEEKING THE SUPPORT OF THE SILENT MAJORITY: ARE LURKING USERS
VALUABLE TO UGC PLATFORMS?
Block 4 – 2022/2023
WEEK 1 – THE SOCIAL MEDIA LANDSCAPE
Social media informs purchasing
- We use Social network Sites (SNS) to research products we want to buy
- We use SNS to discover brands / products we otherwise wouldn’t be aware of
- We use SNS to buy products
User Generated Content (UGC) = content generated or created by an internet user who is a
consumer of this information or content. They’re contributing back to the platform.
Social Media = the online platform that hosts this User-Generated Content.
- With this broad definition, we include many platforms that we traditionally don’t view as
Social Media (e.g. NYTimes, Uber, Rotten Tomatoes, Amazon, StackOverflow and WordPress)
Reasons why Social Media has taken off:
- Digital Social Networks: for a structured information flow around us
- Machine Intelligence: the ease and speed of recommendations they give
- Smartphones: we’re always on, and they send us notifications which we like
Aral’s Hype Machine: the previous 3 factors are layered on top of each other, to facilitate Social
Media growth.
- Because of the Digital Social Networks, we’re digitally connected to each other that’s
observable via social graphs.
- Machine Intelligence takes this information and guides consumers through content they may
want to consume (friends they may want to follow). Which is now doable because:
o Richness of the data: the network is now observable
o Advances in computational power: Smartphones are fast enough to do this
- Which is then used to give us notifications / signals via our Smartphone 24/7. Our brain likes
these signals / reactions we get (likes gives dopamine).
o We also use our smartphone to input data, by writing on Social Media, and thereby
continuously feed need information into the network, which is again used by
machine intelligence to create better recommendations.
Dark side of Social Media:
- Mental health effects: social media is detrimental to our health
- Echo chambers due to self-selection into content and recommender systems
o Different (groups of) individuals do self-selection into what we read. And
recommender system will keep on sending this information tailored to our needs,
that in the end, we do not get a clear broad view of what’s happening.
o This also happens in politics: political polarization (where you as a democrat only see
posts putting democrats in a positive light).
- Privacy concerns: we’re not aware how many data is collected on us online.
Why do people contribute to social media? Why do they want an audience?
- Intrinsic utility: inherent satisfaction
o Users receive direct utility from posting content
o An exogeneous increase in followers, is expected to lead to more posting.
, o Because we’re happy that we’re broadcasting to more people, which is an incentive
to post more.
- Image-related utility: status-seeking / prestige motivation
o We’re motivated by perceptions of others
o An exogeneous increase in followers, is expected to lead to no change or even
decrease in posting behavior
o Because we post to attract followers, and once we have them (a minimum number of
social bonds), it’s enough.
STUDY – TOUBIA & STEPHAN (2015) | INTRINSIC AND IMAGE-RELATED UTILITY
- A field experiment on 2500 twitter accounts
- They used 1335 active users, and gradually added 100 new followers to 100 of these
accounts.
- Results:
o No main effect on positing activity: there was no significant difference between those
100 accounts (treatment) and the other 1235 (control) in posting rate.
o A differential effect based on the number of followers started with
▪ Quintile 2 (out of 5): relatively low # followers leads to an increase in posting
rate (= intrinsic-utility effect)
▪ Quintile 4: relatively high # followers leads to a decrease in posting rate (=
image-related utility)
- Managerial implications:
o Brand advocates aren’t always those with large follower counts
▪ Image-related utility concerns start to come into play
▪ Micro / nano influencers may work better in that case
o As a social media platform matures (getting more and more followers), we see more
firm generated content
▪ Consumers stop broadcasting because of image-related utility
▪ Firms no longer use it as a social listening platform, but an advertising
platform.
Lurker = a user of a social media site who doesn’t actively participate / contribute, but they do
consumer information
- 90-9-1 rule: 90% of users are lurkers, 9% contribute from time to time, 1% account for the
most contributions
- But these numbers aren’t always correct, they change a lot.
- But the exact distribution doesn’t matter, as long as we do take into account that lurking
matters and it has a huge impact.
The marketing value of lurkers (Chen, 2019)
- Lurkers aren’t low value users, they still engage and make decisions based on what they see
online
- There’s a distinction:
o Passive lurkers = absorb content and don’t spread information
▪ They generate value from their own decisions
o Active lurkers (diffusers) = transmit information to others (after reading)
▪ They generate value from
• Their own decisions
• Diffusion: they share information with others offline
STUDY – CHEN (2019) | SEEKING THE SUPPORT OF THE SILENT MAJORITY: ARE LURKING USERS
VALUABLE TO UGC PLATFORMS?
