The Story

I recently bought a new HD TV from “ABC TV Maker” and registered at the point of sale for warranty. Within a week, my satellite provider had phoned to see if I wanted to upgrade to their HD service and also tried to sell me their movie package. They had bought the data on HD TVs and they were using it to target users, perhaps in a certain income bracket, perhaps with a certain family size, who had recently bought a HD TV. They were using data analytics for marketing.

The data analytics

“ABC TV Maker” has recognised the revenue opportunity in selling information on their customers’ behaviour to other businesses. They have sold details of customers who have recently bought a HD TV either directly to the satellite provider, or to an intermediary data provider. The satellite provider has then cross-matched the list of customers who bought a new HD TV from “ABC TV Maker” (and probably other electronics providers) against a list of their own customers. They have then removed from the list any customers who already have a high-value package , and targeted the remaining customers with a phone call. It is possible (and in fact probable) that the data analytics used was more sophisticated than those described above. The data analytics described above involves the following processes:

• Matching of customers’ and prospects’ identity information across possibly disparate and unreliable data sources
• Data integrity checking

I called my ‘mobile operator’ to cancel my subscription recently and I was put through to their ‘retention team’. This team eventually reduced my monthly subscription, and doubled the number of free minutes and free texts. They had had a data analytics tool that gave them the information on which to make the following decisions:

•    Is Pádraig worth retaining as a customer?
•    What offers will entice Pádraig to deliver the most value to us?
•    What is the least profitable offer we are willing to make to Pádraig?

The data analytics

The operator has a variety of data sources available, providing information on almost every consumer in the UK. These data sources are derived from third parties, in addition to their own customer data.

They have used data analytics to maximise the revenue opportunity from each customer contact with their retention team. These data include information such as how old someone is, what types of holidays they like to take, what newspapers they buy, whether they live in an area populated mostly by affluent retired people, or by students, and many others.

They also include statistical scores, or segments, which attempt to condense a large volume of data about each individual into a few highly predictive numbers. An example of one of these numbers is known as a risk score. The risk score attempts to predict how likely each individual in the UK is to fail to pay for the services which they provide. Risk scores often used by companies to decide whether or not to extend credit to a current or potential customer.

The marketing analytics department at the mobile operator have taken all of these data, and used advanced statistical methods to predict the lifetime value of each of their customers, and the type of usage plan and phone they are most likely to desire. By combining this information, again using advanced statistical methods, they push the appropriate offers through to the dashboard of the contact centre worker taking the call from Pádraig.

The result is a win-win. Pádraig gets the price plan he wants at a cost he feels is good value. He wasn’t offered products or plans he wasn’t interested in. It was a quick call, focused on his needs. It’s a great customer experience, and Pádraig is more likely to return to the same operator for a deal in the future. From the mobile operator’s point of view, they sold the most profitable deal to Pádraig that he would be interested in, and now they have a chance to cross and up-sell (more of which later …)

A typical data analytics solution will involve phases of investigation, discovery, understanding, refinement, delivery and formatting of information present in data. These phases will usually be employed within an iterative process.

A well-defined data analytics project is known typically to provide a return on investment in excess of 10:1 on the investment in analytic resources.

Data analytics can be used by forward thinking companies in the areas of operations, marketing, product design, and risk.

• Are you able effectively to control every key aspect of your business?
• If not, then how do you take control? What information do you need, and what drivers need to be addressed?
• Now that you are in control, where are your opportunities?

The methodology within this framework is detailed, and covers business and technological processes. It derives from Agile project management methods and the advanced control science developed for use in political planning and space science. We use it because we have found from our experience serving some of the world’s biggest companies that it brings significant material benefits to clients.

We are first and foremost, however a business company. We talk business solutions, and quietly keep the high-tech analytics where it belongs, working for clients’ benefit. If you’re interested to know more of the high tech, just keep reading our blog: interspersed with our business level posts we’ll present some of the advanced methods in bite sized chunks we hope anyone can swallow in their lunch hour. If you can’t wait then leave us a comment and we’ll get back to you.