Why Are the Largest Social Networking Services Sometimes Unable to Sustain Themselves?

Round 1

Reviewer 1 Report

This is the research for the sustainability of Social Networking Service. This paper analyze it in terms of both academic and business perspectives. This paper is logically written in excellent English. There are lots of cases; KakaoTalk, Cyworld, Myspace and Facebook. This paper describe and analyze them historically and tried to find out the sustainability and life of Social Networking Services. This research is very interesting and valuable for the business strategy of survival Social Networking Services. I recommend the acceptance of publication. 

Author Response

thank you very much for your nice review!

 

Reviewer 2 Report

The work provided is significant, but the way it is presented lacks care.

Here are some remarks to better highlight the work presented:

1. The quality of Figure 2 needs to be improved. The same goes for the Tables 2-7.

2. The title of Table 4 is hidden!

3. Mathematical formulas as Equation of line 324

5. Results commented on lines 452-454, 459-463, 549-551, 555-461 must be edited in a more rigorous format.

6. Apart from a case study approach, the real contribution of the paper is not visible and should be better demonstrated.

Author Response

The quality of Figure 2 needs to be improved. The same goes for the Tables 2-7.

 ==> As you can see in the revision file, We upgraded the quality of Figure2, and from Table 2 to Table 7.

The title of Table 4 is hidden!

 ==> Fixed. ( I worked with Mac computer, so i didnt realize how the formats are broken. I worked again in Window PC now at PC bang. )

Mathematical formulas as Equation of line 324

==> We Put '= ' in the section now.

Results commented on lines 452-454, 459-463, 549-551, 555-461 must be edited in a more rigorous format.

==> We re-formatted all of diagram images(Figure5~12) and also added full subset/largeset notation for large or small relation between paired cases according to a time change .

Apart from a case study approach, the real contribution of the paper is not visible and should be better demonstrated.

==>  We think we contributed at least with 6 major contributing points. let me summarize like below. (also added to last section in revisioned article)

  1)  Most of all, we gave a answer to 'why and how sometimes the largest SNS cannot sustain themselves?'  by building up hypothesis.  As in the Dul and Hak (2008),Hypothesis Building without hypothesis testing has also valuable research method.  2) According to the results of this research, network effect is never dead[33].There has been many articles with doubts on Network Effect these days. In order to sustain the network effect, at least in the domain of Social Networking Service industry (including both Facebook and messenger services), it needs to occupy the most of the offline social network(Larger Population Social Network index bigger than roughly  50%). Paradoxically, the contribution of this study is that the first mover company’s failure despite of the existing network effect, is not because there is no ‘network effect’ anymore, but rather it strengthened the importance of the network effect in more right way especially in online/mobile Social Networking Service industry.

There is 4 more contributing points. First, We can argue API strategy for sustainability of existing SNS is not sufficient condition[13]. Facebook executed the API strategy only after they kept continuing successfully replicating offline social network first.( interweaving local networks to naturally produce global social networks on their platform.). And even if Cyworld once copied the API strategy of Facebook, they weren’t successful because they didn’t keep continuing to accumulate pure offline social network into their existing platform first. Second, Timely accepting mobile or technology paradigm shift can  not be a sufficient condition to sustain existing network effects[14]. Facebook outwit Cyworld and Myspace even before Mobile era fully blew up. Cyworld or Myspace lost their controlship because they didn’t keep continuing accumulating precious offline social network database. This means also they didn’t raise the value of Larger Population Social Network(LPSN) index or depth rate(average number of friends on each SNS per Dunbar Number 150)  not only width rate(# of users of SNS per local real population).

Third, This hypothesis can  be practically utilized for practitioners in SNS industries to decide if (s)he would start new venture to defeat Facebook  or wechat etc, if s(he) could be safe with existing SNS s(he) is operating as #1 position in the industry.  Fourth, In academia, Most of articles are written with assumption that all kinds of platforms are equal to each other in analyzing network effects on each platform. But, There is originally big difference among various kinds of platforms. Amazon (Commerce platform) and Airbnb or Uber (Online to Offline platform) or Facebook or Linkedin (Social Network platfrom) or Pinterest or twitter (content network platform ) have their own existing principles. For example, Users of Amazon or users of Airbnb is not connected to each other.  And Users of pinterest don't have to know each other very well as like Facebook or Linkedin. These means basic structure of each service, and contents type and user relation type are all different which makes the Network Effect phenomenon quite different.  For example, Network Effect comes earlier than  fully being safe from late coming competitors. Myspace or Cyworld had huge hockey stick growth path with Network Effect. But, because they didn't replicate most of offline social network (which are kind of mother set or large set)  enough, they eventually couldn't keep their first mover advantage.

 

 I hope these could be a good answers or works for the reviews.

 

Pls let me know if there is any mistake or something for me to fix quickly.

 

With warm regards,

 

Yong

Author Response File: Author Response.docx

Reviewer 3 Report

There are serious flaws in this paper which are given as under.

The Abstract does not provide the objectives of the study, the problem solved in this study, the methodology used, the technique of the analysis, the study results and what are the implications of this study on the current state of the art.  Why a threshold of 30% used? 'Why depth and width are important' is not explained. The write-up is difficult to follow and understand.  The Abstract is useless. No information about methods, study results, and implications is provided. The tables/figures/charts need a rework. They are very difficult to understand at their current form.  The conclusions section is not supported by the results.  Study implications are missing. The research methodology is not significant.  The introduction section is so short. The discussion section is missing. 

Author Response

I clarified what you pointed out , to the abstract/introduction/conclusion parts.

and here i summarize to each of your points. thanks.

 

-the objective of the study  : To build a hypothesis explaining to why and how the largest SNS collpase even though they were enjoying network effects?

-the methodology used : multiple case study/practice oriented method/hypothesis building (not testing) method /for validity and reliability ,chain of evidence (LPSN function ) and case study protocol setup with data and investigator triangulations.

-study results : found simple function, named LPSN index, the level how much the SNS occupied or copied or replicated offline social network universe or real society's total human networks.

-Implication :

We think we contributed at least with 4 major contributing points and 2 main implications. let me summarize like below. (also added to last section in revisioned article)

  1)  Most of all, we gave a answer to 'why and how sometimes the largest SNS cannot sustain themselves?'  by building up hypothesis.  As in the Dul and Hak (2008),Hypothesis Building without hypothesis testing has also valuable research method.  2) According to the results of this research, network effect is never dead[33].There has been many articles with doubts on Network Effect these days. In order to sustain the network effect, at least in the domain of Social Networking Service industry (including both Facebook and messenger services), it needs to occupy the most of the offline social network(Larger Population Social Network index bigger than roughly  50%). Paradoxically, the contribution of this study is that the first mover company’s failure despite of the existing network effect, is not because there is no ‘network effect’ anymore, but rather it strengthened the importance of the network effect in more right way especially in online/mobile Social Networking Service industry.

There is 2 more contributing points. First, We can argue API strategy for sustainability of existing SNS is not sufficient condition[13]. Facebook executed the API strategy only after they kept continuing successfully replicating offline social network first.( interweaving local networks to naturally produce global social networks on their platform.). And even if Cyworld once copied the API strategy of Facebook, they weren’t successful because they didn’t keep continuing to accumulate pure offline social network into their existing platform first. Second, Timely accepting mobile or technology paradigm shift can  not be a sufficient condition to sustain existing network effects[14]. Facebook outwit Cyworld and Myspace even before Mobile era fully blew up. Cyworld or Myspace lost their controlship because they didn’t keep continuing accumulating precious offline social network database. This means also they didn’t raise the value of Larger Population Social Network(LPSN) index or depth rate(average number of friends on each SNS per Dunbar Number 150)  not only width rate(# of users of SNS per local real population).

Implication 1, This hypothesis can  be practically utilized for practitioners in SNS industries to decide if (s)he would start new venture to defeat Facebook  or wechat etc, if s(he) could be safe with existing SNS s(he) is operating as #1 position in the industry.  Implication 2, In academia, Most of articles are written with assumption that all kinds of platforms are equal to each other in analyzing network effects on each platform. But, There is originally big difference among various kinds of platforms. Amazon (Commerce platform) and Airbnb or Uber (Online to Offline platform) or Facebook or Linkedin (Social Network platfrom) or Pinterest or twitter (content network platform ) have their own existing principles. For example, Users of Amazon or users of Airbnb is not connected to each other.  And Users of pinterest don't have to know each other very well as like Facebook or Linkedin. These means basic structure of each service, and contents type and user relation type are all different which makes the Network Effect phenomenon quite different.  For example, Network Effect comes earlier than  fully being safe from late coming competitors. Myspace or Cyworld had huge hockey stick growth path with Network Effect. But, because they didn't replicate most of offline social network (which are kind of mother set or large set)  enough, they eventually couldn't keep their first mover advantage.

- Why a threshold of 30% used? : I would not argue with this. I just put the 30% because all of cases who lost their position to latecomer, their LPSN index were far lower than 30%.  I omitted this part.

-'Why depth and width are important' is not explained.: A SNS's replication level of offline real social network (person to person's relationship data or node and link data between people in real world.) is regardes as main driving force for latecomer to outwit first runner. So, we had to make formula to calculate the replication level or rate. And for that purpose, width rate and depth rate seemed good fit considering cost and benefit of getting the data.

 

-The tables/figures/charts need a rework. : all are reworked. thanks.

- the results / Study implications are clarified at the conclusion section.

-The discussion section is missing : added discussion chapter independently right before conclusion section.

pls let me know if i misunderstood to what you pointed out. thank you very much.

Author Response File: Author Response.docx

Reviewer 4 Report

Why sometimes the largest SNS cannot sustain themselves?

This study aims to answer a research question of “why sometimes the largest SNS cannot sustain the No.1 positions?” even with the network effect. In other words, “when (in what condition) does the largest SNS collapse?” This study answers this question with four cases, that is, NateOn Messenger catching up #1 Messenger MSN in Korea (Case 1), KakaoTalk catching up #1 Messenger NateOn in Korea (Case 2), Facebook catching up Myspace in USA (Case 3), and Facebook catching up #1 Social Networking Service Cyworld in Korea (Case 4).

As a result, this study argues “a largest social networking service can collapse when a new social networking service is growing in another larger population social network (LPSN). Or, LPSN index lower than 30% can be almost very dangerous to lose its controlship in the winning market.”

 

First of all, the topic is very interesting and important in the current, pervasively digitized world, in particular, where digital platforms become the key strategic way to achieve competitive advantage and online social network service firms become powerful market players. The research question is interesting because it defies the network effect in online social network service. The four qualitative case studies are interesting and useful because the cases capture not just normal but exceptional phenomena where dominant players were defeated by new comers and thus we can effectively find the answers for the research question. Thus, the methods are also well designed and aligned with the topic.

 

In general, the paper is interesting and has a great potential to make an important contribution, and I do not find a critical issue in this paper that cannot be fixed with a careful revision, except for some minor suggestions, which I explain in the following sentences. I hope my comments help the authors further polish and streamline the story.

 

First, I recommend the authors to define more clearly and specifically “Larger Population Social Network (LPSN)” which is currently defined as the social network with population larger than what existing social networking services aims occupy. This is not clear enough to support the suggested theoretical hypothesis and arguments with findings from case studies. For example, what is the LPSN in each of the cases? All the SNS firms might have the same LPSN but it seems not described that way. Is the LPSN in all four cases PC vs. Mobile users? It seems like the authors explain the population of Korea or US as LPSN or the world population as LPSN. If the definition of LPSN is explained more clearly and specifically, this ambiguity could be resolved.   

Scone, some parts of the paper could be better understood with more detailed explanation. For example, more detailed explanation of all tables and figures, especially Table 4~8. Figure 5~10 could be better explained if the LPSN definition is updated based on my comment said above.  

 

Lastly, please explain why you choose LPSN index 30% as a threshold for sustaining SNS leadership.

 

Some minor comments: If you did not make Figure 1 but just copied from Dul and Hak (2007), then it should be acknowledged in Figure. Otherwise, you may clearly explain you made it.

 

Please double check grammar and typos which I can find in multiple parts (e.g., page 14: 355, 359, Figure 3 and Figure 5 should be Figure 5 and Figure 6). Fix several typos also.

 

Best of Luck!

 

 

 

Author Response

I inserted  changes  into the document, and below is brief summary point by points as to the reviewer's comments.  This reviewer seems very friendly and sophisticated ,so very helpful ! thanks again! 

  ----------------------------------------------------------------
- 30% ==> omitted 30% itself, because so many controversy on this.  rather changed paragraphs like this.  “ If LPSN index (width rate* depth rate) of early starter becomes lower than latecomer’s index, latecomer can easily overtake the first position   -  Definition of Larger Population Social Network is reinforced like this.   ==>. The largest population social network is the offline social network universe itself. And in one person’s view, the larger population social network exists in his(her) brain than in offline rolodex or email or cellphone number in digital formats. So the size comparison  of Larger Population Social Network can be represented by sign of inequalities as like below.

 

Offline Social Network >  one person’s brain memory about people who he(she) once met > rolodex or digital addressbook he(she) keeps. 

 

So, Larger Population Social Network not only consider population of one country or world , but also relation information between two selected persons. In other words, It can be meant by degree or depth of replication of offline node and links(every person’s nodes and links means social network) in this society altogether into specific container(offline or online; hardware rolodex or software keeping email list or friends list or smartphone or future VR/AR Smart Glasses etc.). So, even if mobile space or container can include more broader or bigger population of one country or world, that is not enough to create Larger Population Social Network. That also requires links or connection information between two persons altogether to create Larger Population Social Network. As we saw cases above, Facebook outwits Cyworld or Myspace even before Mobile Space era fully arrived, because Facebook kept fetching offline social network by dragging not only nodes but also links between them, whereas Cyworld and Myspace were negligent in doing same efforts. And NateOn fetched offline social network by utilizing existing nodes and links database(LPSN) already created by Cyworld and Cellular Network of SKT. And KakaoTalk fetched offline social network by utilizing existing nodes and links database(LPSN) already created on smartphone phone addressbook.

 

 

==> Book of Dul and Hak (2008) added on reference.

==> Figure typos fixed.(45.2. NateOn vs. KakaoTalk)   

 

Best Regards,

 

Yong

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I would like to thank the authors for taking into consideration all the remarks I made during the first round review.

I therefore consider that the answers as well as the improvements made allow me to recommend this article for publication in Sustainability Journal.

Author Response

thank you very much for all of your efforts and consideration !  thank you ~~

Reviewer 3 Report

The authors have improved it much as per my and the suggestions of the other reviewers. However, there are still some issues that authors need to address:

The Introduction section is still so short. I suggest at least add two to three more paragraphs in it. The use of the English language is still problematic. It can only be fixed by an extensive language edit by a native speaker of the English language who is also an expert in this field.  

Author Response

The Introduction section is still so short. I suggest at least add two to three more paragraphs in it.

 

==> Added two more paragraphs in it. 

The use of the English language is still problematic. It can only be fixed by an extensive language edit by a native speaker of the English language who is also an expert in this field

==> extensive language editted with my Indian fellows who are experts at blockchain based social networking service.

 

I attached a revisioned file.

 

thank you very much.

 

Best Regards,

 

Yong

Author Response File: Author Response.docx