CASE STUDY: Oh, My Darling, Data Mining!

June 26th, 2006:Direct marketing is a tough game. Leads rapidly expire, prospects peter out and connecting to the right customer can be as elusive as the white stag. With the right solution and vendor, however, you can be sure that your quiver will always be full.

Case Study

If you're a direct marketer, it's not enough to wildly fill letterboxes with mail that will undoubtedly be seen as junk by those who aren't interested. Being able to identify the right kind of prospects at the right time is what drives successful marketing companies forward and not only saves time, energy and expense but also reputation.

Processing over 28 million pieces of mail each year, Virid is one of Australia's largest direct mail marketing companies. And being one of the largest, it required powerful but accessible predictive analytic software to improve its business. It found this in SPSS' Clementine.

Virid made the switch in late 2003 and has neverlooked back saying that Clementine helped them halve mail costs and double campaign profitability in one year.It has helped more informed and strategic decision making, improved staff efficiency and enabled the company to become more agile overall.

No Headlamp, Pick or Database Skills Required
Software is getting more and more powerful, however in order to remain useful it also needs to be user friendly. This means helping staff do their job more efficiently without forcing them to learn a whole new system. After all, you don't want your prize marketing people wasting precious hours learning new database commands when they could be canvassing new prospects right? This is where SPSS's data mining and predictive analysis software Clementine comes in.

Originally, Virid was using MSSQL to dig through its data. A tedious and confronting process for people from less IT-orientated backgrounds.

"Our data-mining staff come from a market research/ marketing area and SQL scripting can be a challenge for people without an IT background" says Sorena Donald, Virid's Market Research Manager. "As such, we needed a product that offered true data mining capabilities and could work with large quantities of data."

Virid looked at all of the major data mining products on the market eventually deciding on Clementine due to its ability to build predictive models, process large quantities of data in short periods of time and plug into a range ofdifferent databases.

According to Donald, its simplicity was also a big selling point as "a broader range of people in the organisation could get involved in data mining.

"When using SQL, you are really limited to people from an IT background. With Clementine, even staff from a marketing background can get involved" she tells IDM.

Clementine has a number of advanced data mining features including built in shortcuts for commonly used data mining functions. This means you save time on complex problem solving and can spend more time getting results.

No More Gut Feelings
Once those results were in Virid was able to use Clementine to build predictive models, traditionally a very complex procedure. It enabled the company to rank external lists from most to least profitable by determining the highest customer profitability from a list over a 12-month period rather than the largest response per campaign.

This had profound implications in the long run, as it brought greater stability to data that could – and had proven in the past – to be fundamentally fickle.

"Its ability to plug into many different databases was also an important factor for us as client data comes in many different forms", comments Donald. With Clementine, existing customer databases can be to be segmented and important information extracted.

Thanks to SPSS' training program, Virid also found that Clementine's ease of use wasn't just limited to its modelling capabilities.

Clementine enables detailed information extraction from a range of databases

"Clementine was easy to learn; it's very intuitive and simplifies a lot of the complexity involved in data mining. The SPSS training was great – it allowed staff to get an overview and hands on experience using Clementine for data mining and modelling. We also got to keep the training manuals and data so there was opportunity to revise at any point," says Donald.

"After the training, we began converting our existing reports into Clementine. We found this process very useful because it enabled us to see the similarities and differences between the two applications and the learning process was faster as a direct result of this."

Once staff were up to speed and comfortable with the new software, Virid saw immediate workflow improvements. However, in the longer term it also noticed even more significant benefits appear.

"In the short term, the most notable gain was in productivity as Clementine enabled us to quickly and easily get the information we needed to make many improvements to the business," says Donald. "Those improvements lead to a significant reduction in wasted mail costs, and improved profitability over 12 months. The improvements in profit allowed us to break even on the investment of Clementine in nine months."

Virid is now sending less mail to better prospects and making more informed decisions overall, drastically improving the way it does business.

"Primarily, it's improved our decision making capabilities. Having the ability to answer business questions as they arise and making the right decisions is vital to the success of an organisation. Specifically, it has enabled us to concentrate our efforts in the areas that provide the highest ROI. Over time, these efforts have made a significant improvement in our profitability" said Donald.

The more we move into an information economy, the more important software like Clementine becomes. SPSS cites cases of up to 600 to 1 improvement on returns during the first year of Clementine use. This may sound ludicrous, however making better use of you intellectual property, streamlining CRM and cutting staff time spent on non-essential processes all compound in the longer term.

Virid found this, and used Clementine's predictive analysis to streamline its business processes, reduce wasted resources and give its campaigns an unprecedented level of accuracy and stability.