Case:Winning back Ex customers forms a very important part of customer expansion strategy for the planning team. Ex customers refers to those customers who had a policy with n the past but are no longer associated with the organization. The campaign planning team sends mails every month to customers whose motor renewal date is due. Since targeting the entire database entails mailing cost and is not feasible, the planning team wants a mechanism which will help it, reduce mailing cost by identify only those customers who will respond favorably to future campaigns

Challenges:Campaign planning team wishes to identify those ex customers who have a high propensity to respond to a motor campaign; however since identifying high propensity customers in a huge database is a complex task and manually impossible team wanted support to capture high propensity customers .

Benefits:

•The model built by us provides an improved quote response rate of 10.8% as compared to pre modeling response rate of 8.01%.Also the cost per quote has reduced from 3.75 to 2.78 p after building the model.

•Without the model customer would have sent 1598 extra volume of mails.

•The extra mailing cost that customer would incur Without the model is 479 pounds

•With the help of the model customer was able to cut down on volumes mailed by 3155.Thus the cost saved by non mailing is 946.4 pounds.

Solution:

To generate an improvement in response over BAU activity,

•Our team developed a robust statistical model using the characteristics of historical customers.

•The Response model thus developed will be used by the planning team to score future customers and only those customers with a higher than average propensity to respond to motor policy will be targeted month over month by the campaign planning team.

Business problem:

• The client is a US-based insurer seeking to improve its customer retention rate by aligning its customer contact strategy with attrition indicators
• Client was aware of the indicators; however they wanted insights into what is the duration post these events after which policy holders actually surrender their policies.

Solution:

The team analyzed 3 indicators and duration after which policy holders actually surrender their policy. Those indicators are:

1. Customer requesting to surrender
2. Request for Illustration
3. Request for duplicate policy

• Gathered and analyzed data using comprehensive quality and reasonability checks and came up with following findings
• a policy holder requesting to surrender does so very quickly
• a policy holder requesting illustration has a longer time period before he/she actually surrenders the policy
• a policy holder requesting dup policy has a longer time period before he/she actually surrenders the policy

Benefits:

The client used this analysis to align its customer contact strategy with the occurrence of these events and days they have to contact these customers post these events. This helped in enhanced retention rates

Objective :

•Client works with retail data to run campaigns on FMCG products by means of various channels

•To use UNICA Campaign as a campaign management tool across all channels

•To migrate campaigns from SQL to UNICA platform.

Process:

•Following steps used for migration

1.Develop a flowchart to segment customers

2.Test this flowchart to match results with existing process

3.Optimize flowchart not to overload or impact the systems

4.Test post campaign analysis and match results with existing process

5.Soft Launch – Run a few campaigns per cycle on UNICA

6.Complete Live – Run all campaigns on UNICA

Impact:

•To segment customers for all campaigns on UNICA

•Migrations led to productivity benefit by reduction in data entry

Business Challenges:

•Executing a trigger campaign for restaurant chain, segmentation from 3 of its sub brands are combined into one flow chart to create a master base table

•For these campaigns, the count of offer segments for each sponsor range between 8-10 resulting in overall 24-20 segments

• Query time for final master table is very high ( approx 8 hours*) since Demographics & Transactions table is accessed by large no of offer segments

Improvement & Execution :

•Extract’ Process box in Affinium Flowchart was used for each sponsor segmentation to create a temporary table consisting of relevant fields only ( and not the entire Demographics & Accounts data)

•Using this option, Query running time was considerably reduced as offer segments had to access the much smaller sized Temp tables created from extract process box, instead of accessing demographics & Accounts table

Impacts/Benefits :

•Base Table Creation Cycle time reduction by 62%

•Significant Productivity savings and Campaign cycle time reduction

Challenges :Business was looking into one source of truth for theirdigital Campaign performance monitoring strategy on WW level which are allign with their Markeing & Business priorities in both PC and Print business.

Step:

• Created framework agrred by BU leadand filtered out important KPI which was allgined with particular business priorities for particualr BU on World wide level(WW)
• Cosolidated all camapign measurement reports and understand metric selection process
• Educate regional lead on KPIs and their importance
• Created approved framework forWeekly/Monthly/Quarterly manner.

Benfits : Data driven QBR with Sr leadership team which helps BU lead and regional lead to take proactive approach intheir digital campaign strategy which imporved their productivity and helped them in resourceallocation and buget management .