As part of the Swiss CRM Expert Forum, the insurance company Suva invited experts and entrepreneurs to Lucerne on April 12, 2018 to talk about "customer data and profiles", exchange experiences and present solutions. Initiator and founder of the forum Martin Stadelmann , Managing Director of ec4u expert consulting, led the day and moderated the exciting discussions.
Jan Lundak, Head of Partner Data Management at SUVA, presented the Data Quality Management of his company. He summarized his talk for our blog.
The initial situation: poor data ensure bad user experiences
A clean database is the foundation for companies to satisfy partners, customers and employees. But a database is not set in stone. New data is constantly flowing in, while old data is being replaced, deleted or combined. This can lead to the construction of a number of construction sites over time, which may result in outdated, incomplete, incorrect or duplicate data, making processing considerably more difficult.
In 2015, we had the problem in our company that we had massive costs and / or resource expenditures due to erroneous data sets in different areas. While certain areas created problems through mass processing of data, other people were busy merging, correcting, and maintaining the data to make it usable. As a result of inaccurate (address) data, there were high costs for postal returns and even higher costs due to time-consuming manual correction processes.
In the meantime, a digitally-planned information campaign had to be redirected to the post office, as we found out that we only had a usable e-mail address for our contacts in only 2.4% of the cases. A personalized e-mail campaign was impossible.
Even the telephone contact was made difficult at this time by various data quality problems. On average, one-third of our contact groups had problems identifying customer contact details directly from the phone number because the stored numbers were outdated or missing.
In short:
A clean database is the foundation for companies to satisfy partners, customers and employees. But a database is not set in stone. New data is constantly flowing in, while old data is being replaced, deleted or combined. This can lead to the construction of a number of construction sites over time, which may result in outdated, incomplete, incorrect or duplicate data, making processing considerably more difficult.
In 2015, we had the problem in our company that we had massive costs and / or resource expenditures due to erroneous data sets in different areas. While certain areas created problems through mass processing of data, other people were busy merging, correcting, and maintaining the data to make it usable. As a result of inaccurate (address) data, there were high costs for postal returns and even higher costs due to time-consuming manual correction processes.
In the meantime, a digitally-planned information campaign had to be redirected to the post office, as we found out that we only had a usable e-mail address for our contacts in only 2.4% of the cases. A personalized e-mail campaign was impossible.
Even the telephone contact was made difficult at this time by various data quality problems. On average, one-third of our contact groups had problems identifying customer contact details directly from the phone number because the stored numbers were outdated or missing.
In short:
The (inadequate) quality of our database made our staff resource intensive whenever they wanted to review or use contact information.
Sometimes our customers and partners got too much, no or wrong information because the contact details were incorrect or missing.
And we were not able to design our marketing and communication strategies as modern, digital and, above all, customer-centered as we wanted, because the database regularly thwarted our calculations.
Before we could come up with a solution, we first had to identify why our database did not meet our expectations. In our analysis we identified three main points.
Before we could come up with a solution, we first had to identify why our database did not meet our expectations. In our analysis we identified three main points.
The causes: contaminated sites, lack of discipline and silos
Contaminated Sites
The longer a database exists, the higher the likelihood of outdated data. Often these are simply migrated with the implementation of new systems and grow over time.
The clean-up of contaminated sites is meanwhile extremely difficult and expensive. Systemic dependencies often prevent a simple cleanup.
Lack of discipline
Despite frequent contacts with the customer by various employees from various specialist processes, there has hitherto been little discipline in updating and adding new customer data in the central CRM system provided for this purpose.
In addition, the customer data is often scattered over many systems and sometimes even difficult to find in Excel files, which brings me to point 3.
Silo projects
Due to the strong focus on the needs of their own specialist processes and the processing of their own business processes, projects are often implemented in the specialist silos . The highly focused implementation neglects overarching and sustainable data quality interests.
The logical consequence of these silos are so-called "island systems". Although these systems are logically interconnected by communication interfaces, data-quality data design is lacking. A fault-tolerant application design is therefore the order of the day.
Based on these findings, we developed a concept for CRM partner data management and processes in 2015 and were able to create a position for Data Quality Management or Partner Data Management by the end of the year. This feature validates data quality, develops and implements optimization and cleanup solutions, and measures and evaluates progress.
Contaminated Sites
The longer a database exists, the higher the likelihood of outdated data. Often these are simply migrated with the implementation of new systems and grow over time.
The clean-up of contaminated sites is meanwhile extremely difficult and expensive. Systemic dependencies often prevent a simple cleanup.
Lack of discipline
Despite frequent contacts with the customer by various employees from various specialist processes, there has hitherto been little discipline in updating and adding new customer data in the central CRM system provided for this purpose.
In addition, the customer data is often scattered over many systems and sometimes even difficult to find in Excel files, which brings me to point 3.
Silo projects
Due to the strong focus on the needs of their own specialist processes and the processing of their own business processes, projects are often implemented in the specialist silos . The highly focused implementation neglects overarching and sustainable data quality interests.
The logical consequence of these silos are so-called "island systems". Although these systems are logically interconnected by communication interfaces, data-quality data design is lacking. A fault-tolerant application design is therefore the order of the day.
Based on these findings, we developed a concept for CRM partner data management and processes in 2015 and were able to create a position for Data Quality Management or Partner Data Management by the end of the year. This feature validates data quality, develops and implements optimization and cleanup solutions, and measures and evaluates progress.
The solution: technology, standards and rules
Technology:
For adjustment and regular control as well as enrichment we have commissioned an external service provider who supports us mechanically and manually.
In order to introduce scattered data into the new system, especially in Excel lists, we have used a CRM data importer. Contacts that appear on these lists can be read directly into CRM, including duplicate detection.
Among other things, our employees are equipped with mobile devices for field service in order to gain access to our CRM system. In customer service, iFY telephony helps to match phone numbers in real-time with existing customer data, and to call up the appropriate contact on the screen. Missing data is displayed immediately, so that the service employee can ask and supplement if necessary.
standards:
Uniform collection guidelines will help to control and regulate decentralized data collection in the future. In addition, it was necessary to adapt all specialist processes and AKVs (tasks, competences, responsibilities) in order to establish and institutionalize data quality-relevant processes.
A data quality index developed by us measures our master data on a monthly basis and generates a report. Based on this, annual data quality goals can be set and verified.
For adjustment and regular control as well as enrichment we have commissioned an external service provider who supports us mechanically and manually.
In order to introduce scattered data into the new system, especially in Excel lists, we have used a CRM data importer. Contacts that appear on these lists can be read directly into CRM, including duplicate detection.
Among other things, our employees are equipped with mobile devices for field service in order to gain access to our CRM system. In customer service, iFY telephony helps to match phone numbers in real-time with existing customer data, and to call up the appropriate contact on the screen. Missing data is displayed immediately, so that the service employee can ask and supplement if necessary.
standards:
Uniform collection guidelines will help to control and regulate decentralized data collection in the future. In addition, it was necessary to adapt all specialist processes and AKVs (tasks, competences, responsibilities) in order to establish and institutionalize data quality-relevant processes.
A data quality index developed by us measures our master data on a monthly basis and generates a report. Based on this, annual data quality goals can be set and verified.
DOS and don'ts:
Before each project, the question must be answered, whether master data is used and how. The concept must then be submitted to the partner data manager for review.
In addition, we want to greatly restrict the manual processing of data in the future. Also, the control when creating new records, whether this may already exist in the system, should be strengthened to avoid duplication.
In addition, we want to motivate our employees to use interfaces to customer data more frequently to update contact information.
Before each project, the question must be answered, whether master data is used and how. The concept must then be submitted to the partner data manager for review.
In addition, we want to greatly restrict the manual processing of data in the future. Also, the control when creating new records, whether this may already exist in the system, should be strengthened to avoid duplication.
In addition, we want to motivate our employees to use interfaces to customer data more frequently to update contact information.
Conclusion:
With clear definitions of master data attributes and collection policies that can be viewed by employees at all times, data quality can be better met in the future. This is supported by monthly measurements and analyzes and data quality goals. Technological support is provided by applications and systems that make it easier to verify, use, and clean data, while preventing / complicating the entry of incorrect records.
Regular data cleansing including a catalog of measures also ensures that the planned measures can be continuously reviewed and, if necessary, strengthened or adjusted.
Of course, these measures and changes require support from top management and sponsors within the company that motivate and inform. Every user has to be involved in the process and changes. Although anxiety is a driving factor, it should probably be used deliberately. It is much more important to be patient. In addition to the technological changes, data quality management also requires a change in processes and the handling of data. This requires time, coaching and patience.
With clear definitions of master data attributes and collection policies that can be viewed by employees at all times, data quality can be better met in the future. This is supported by monthly measurements and analyzes and data quality goals. Technological support is provided by applications and systems that make it easier to verify, use, and clean data, while preventing / complicating the entry of incorrect records.
Regular data cleansing including a catalog of measures also ensures that the planned measures can be continuously reviewed and, if necessary, strengthened or adjusted.
Of course, these measures and changes require support from top management and sponsors within the company that motivate and inform. Every user has to be involved in the process and changes. Although anxiety is a driving factor, it should probably be used deliberately. It is much more important to be patient. In addition to the technological changes, data quality management also requires a change in processes and the handling of data. This requires time, coaching and patience.
0 Comments