Social Call Behaviour Analytics At Optus

[Steve Bennett]

There's an interesting podcast discussing 'Social Networking Analysis at Optus' that you can listen to here:

This is from Teradata's in-house iQ magazine in Australia and their synopsis of the podcast is:

"Tim Manns from Optus, discusses how the company uses detailed network data from its Teradata system to look at calling behaviour. With 40% of the Australian telecommunications market, the company cross-references each customer with every other customer, groups them together based on who they communicate with, looks at the behaviour of the group, and can then predict next steps and target those groups with appropriate products and services."

Although this is a new initiative for Optus, the interesting information is in the discussion about how Optus is grouping their customers to predict their collective behaviour in regard to churn and buying. Optus has seen that if one customer leaves Optus, then one of the rest of the group is 4 or 5 times more likely to also leave. Knowing this, the CRM can trigger an event to start a marketing activity to retain the group members

In terms of marketing, Optus uses these groups to market offers that include 'circle of friends' that suggest a list of the friends the subscriber may want to include.

Getting down to the detail, Tim Manns talks about how Teradata is used to process (parse and clean telephone numbers) 7 days of data (550 million rows) in about 30 minutes. This is then matched to additional data such as customer spend, address and other call details.

Interesting stuff and credit to Teradata for organising a local case study.

More...