I'm quite proud of the social network analysis (SNA) that I'd first completed months ago. It is refreshed each month (the data warehouse load is too high to run it daily or weekly as I would like). I've been tracking its performance, and am continually surprised.
The trouble is that my colleagues are having trouble understanding how they can use it to formalise customer communications, so I decided to try a different approach than graphs and piecharts etc.
Instead I thought I might try something humorous, hence Dilbert to the rescue! I have created a dozen or so custom Dilbert slides that provides some info about a customer insight made available by the SNA and also has a humorous conclusion to those insights. I'll pass this around the department in a series of daily emails.
Here is one example (I had to change the project nickname to "SNA" for this blog);
9 comments:
... ?
FAIL.
:)
I though it was a novel idea to present some analysis to business users without numbers. I'll let you know how it goes.
Hi Tim
SNA sounds really interesting - do you have some presentation material you could share, perhaps the information you presented at the recent conference? I would be especially interested if you have used Clementine in undertaking the analysis.
Dominic.
Hello Dominic,
Yes, I did (and do) use SPSS Clementine for most of my work. The SNA work I've done involves a dozen or so separate steps, but the whole thing runs from one Clementine stream file.
There is a little bit of custom SQL embedded in three source nodes, so that a few joins are more optimised using hard coded dates (very important when you're working with billions of rows), but most of the analysis is relatively simple Clementine streams.
Unfortunately it is difficult to discuss without losing competitive advantage, but I did write some info in a previous post;
SNA In Mobile NetworksTeradata was interested in the work Optus were doing and asked me to discuss in a podcast our data mining analysis (running on Teradata). I managed to get approval and am waiting on Teradata to let me know when the podcast is publicly available. It mentions SNA.
Cheers
Tim
Gotta love Dilbert! I have actually seen some data mining Dilbert comics that were published. I guess with complicated techniques it is always going to be hard to convey the message. Hopefully though, the advantages will speak directly to the KPIs of the people you need on board? Probably the best way to prove the strength of it is to run a test... face off against a control group or the existing models... or is the problem getting the support to do some tests?
Hi Shane,
Yes, the trouble is getting enthusiasm in the first place to run campaigns. I’ve examined historical data and quantified the effect that social networks has for a number of different scenario’s (acquisition, purchase of new handset (iphone), churn etc) but no control group results yet.
In terms of stirring up interest the Dilbert mock-ups have been a huge success, and hopefully very soon we’ll have dozens or new marketing initiatives off the back of this analysis.
Thanks again for the page rank idea. It has proved very useful.
Tim
Hi Tim
I'm just getting into SNA. Do you know of any good post-grad level courses in Australia?
Hi FrancisB,
Nope.
There are some books and blogs that discuss SNA. A book by Carlos Andre Reis Pinheiro is about SNA in telco, and I got sent a copy recently. I'll be reading it when I get some free time.
Personally I try to learn through reading my peers work, and of course continue to learn through trial and error.
Tim
Post a Comment