Sparkasse Zollernalb
How a regional German savings bank used DetectX® predictive analytics to target gold-card upgrades with precision, converting almost 20 percent of the identified customers within six weeks against historical campaign rates of 2 to 4 percent.
A sales campaign, scored before it was sent
This is sales work rather than compliance work, and the page treats it as such. The mechanism is the same predictive analytics platform the compliance modules run on, applied to a question about the bank's own customers.
At a glance
- Sector
- Banking
- Location
- Germany
- Solution
- DetectX® Sales and Marketing
- Website
- sparkasse-zollernalb.de
The story, as Prospero publishes it
About Sparkasse Zollernalb
A public-law savings bank, founded in 1836
- Sparkasse Zollernalb operates as a public-law credit institution owned by the Zollernalb district, affiliated with the German Savings Banks and Giro Association, and functions as the market leader in its region. It maintains 50 branches and self-service locations with 699 employees. As part of the Sparkasse-Finanz group it offers universal banking services including building savings contracts, investment funds, insurance products and leasing arrangements through partner organisations.
Starting position
Precision targeting instead of broad outreach
- The bank defined credit-card sales objectives focused on converting standard cardholders to premium gold-card customers. Rather than addressing all standard cardholders with a high waste rate, management wanted only the customers with real upgrade potential. Anonymous data covering master records, payment transaction data, turnover data and portfolio details for 12,500 standard credit-card holders was provided as the basis for the scoring model.
Method
Weight the attributes, find the pattern, score the customer
- Using DetectX®, the attributes of all data sets were weighted in model calculations to derive patterns relating to purchasing behaviour and the probability of a transaction. Once the relevant patterns had been identified, the customer data was scored against them and divided into potential classes, surfacing the cardholders with the highest upgrade potential. In August the targeted cardholders received a direct-mail offer. The source notes that the tool also takes the results of campaigns already run into account, so its analysis improves with each one.
Result
Almost 20 percent in six weeks, against 2 to 4 percent before
- Almost 20 percent of the customers identified in the scoring process opted for the gold card within six weeks, against historical success rates of between 2 and 4 percent in comparable campaigns run without DetectX®. The source describes a double cost reduction, through careful use of sales resources and through smaller campaign volumes, and a targeted approach with tailored offers leading to greater customer satisfaction. The bank subsequently applied the same method to investment products, pension offerings and lending, and reports over 70 percent accuracy in identifying customers at risk of churn.
What Sparkasse Zollernalb also said
- Birgit Schön, Marketing Private Customers, Sparkasse Zollernalb
- The result with the sales project with Prospero is positive. Based on the Prospero potential scores we achieve a significantly higher contribution margin in the card business.
What the campaign returned
Two published results and the size of the data set behind them, carried exactly as Prospero states them. Neither result is dated and neither is accompanied by absolute numbers.
Published result
Almost 20% conversion
Of the cardholders the scoring identified, almost 20 percent took the gold card within six weeks, against 2 to 4 percent in comparable campaigns without DetectX®.
Over 70% churn accuracy
Accuracy in identifying customers at risk of leaving, on the follow-on work across investment, pension and lending products. No method or sample is published.
12,500 cardholders
The size of the anonymous data set the scoring model was built on: master records, payment transactions, turnover and portfolio details.
The result of the campaign surprised and delighted us. Even after deducting all expenses for data analysis and the campaign, the result of the upgrading campaign now achieves a significantly higher contribution margin in the credit card business.
The contribution margin is described as significantly higher in both of the bank's quotations and is quantified in neither. The conversion figures are Prospero's own, undated, and should not be read as a benchmark.
Where to look next
The mechanism underneath the campaign, the sector page behind it, and the other banking stories.
- Profiling and scoring
- How a score is built, and what moves it. The same mechanism, read for risk rather than for sales.
- Banking
- Why banks choose DetectX®, including the customer intelligence argument this study belongs to.
- Finnova Analytical Framework
- The same predictive analytics platform, embedded in banking software.
Prospero also publishes this case study as a PDF. That file is not held in this site's assets, so there is no download here yet, and nothing above is taken from it.