***** To join INSNA, visit http://www.insna.org *****
Not having the details of the paper, I think the topic can be generalized.
You seem to have chosen phone calls to define the ties between actors in
your network, and customer-level revenue as an actor attribute.
The research could be interesting for a broader audience than telecoms if
the tie definition were left more open (could be giving purchase advice,
forwarding e-mails, sharing online photos etc.) and the revenue per customer
notion extended to a sort of "customer value/net worth over a period of
time" measure, which most serious companies try to keep track of.
If the objective is to understand and come up with ways to optimize the net
worth of the customer or, by extension, of the whole network, a broader tie
definition would make the topic more general by placing the focus on the
methods that you employed to analyze the network rather than on the specific
telecom dataset you used.
Alin
On 12/28/06, Haenlein.Michael <[log in to unmask]> wrote:
>
> ***** To join INSNA, visit http://www.insna.org *****
>
> Dear colleagues,
>
> I'm currently working on a research paper that investigates the
> relationship between revenue generated by people within the same social
> network. For this, I collaborated with a mobile phone company to sample the
> social network of a set of customers. I took a random sample of the customer
> base (about 300 names) and looked at all people called by these customers. I
> then identified people called by these people and called by these people and
> so on ... until I sampled a network consisting of roughly 7,000 actors.
> Hence, essentially what I did is sampling 300 ego-centered networks that,
> after some iterations, started to partly overlap. In a last step I then
> analyzed how revenue generated by one customer can be explained by revenue
> generated by the customers this person is friends with using traditional
> regression analysis.
>
> After submission of my work to an Academic Journal, I received the
> reviewer comment that my analysis is too idiosyncratic for the
> telecommunication industry. Essentially, the reviewer said that only telecom
> companies have access to actual communication patterns of their customers
> and, hence, my analysis could not be replicated in any other industry.
>
> I'm rather new to social network analysis, so I'm a bit unsure whether
> this is true or not. Can anyone think of other industries where a company
> has, at the same time, information about (a) actual communication patterns
> of their customers and (b) customer-level revenue to perform a similar
> analysis as the one I did? Are there any advances in information technology
> that might make such data available in future? Or is my analysis really
> specific to one sector only and will remain so in medium-term?
>
> Any comment would be much appreciated!
> Thanks very much for your help in advance,
>
> Michael
>
> Michael Haenlein
> Professor of Marketing
> ESCP-EAP European School of Management
> 79, Avenue de la République | 75011 Paris | France
>
>
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
>
_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.
|