Planning Information Quality
As with handling raw materials, solid planning and a lot of hard work is required for managing the quality of the customer data asset, companies rely on establishing a framework, which will include an action plan, for data quality initiatives, just as companies have for all other efforts thus far. We will look at five steps in the approach for data quality assurance:
Table of Content
Detail the characteristics of the future quality state (target)
We are aware that data maintenance is costly and that not all customers are the same in this regard. Customers who are more valuable than others should be rewarded with more money and better-quality targets. The reverse is also true.
Customers who perform really well may have accuracy rates of 95 percent or greater. The lowest-value customers may be so inaccurate that companies cannot afford to maintain them any more accurate than 60 to 70 percent of the time, if not even less. If the data quality becomes too poor and companies have no idea who their customers are, the company is not engaged in relational marketing activities.
XYZ identified three global customer segments: the named segment, the small to medium company sector, and the consumer segment. Because XYZ believes that they should be handled as independent groupings and distinguished from one another based on their existing or projected value, these were chosen as a result. Due to the fact that the consequences of making a mistake are significantly bigger for the higher value sectors, most firms establish separate quality objectives for each segment.
For each segment, XYZ defined a new set of data quality objectives as well as a different set of maintenance duties. Without a doubt, the Valencia project was designed to serve a certain consumer niche, and XYZ intended to devote quality maintenance resources exclusively to that particular group of customers.
Detail the characteristics of the current quality state (baseline)
As with all other indicators, trends in data quality are usually more essential than absolute scores, and this is true for all of them. companies can only quantify the change in data quality if the company knows what the baseline quality was for each segment in the first place.
Data quality evaluation tools and analysis software can be used to determine the overall quality of the data. Following the identification of changes in quality levels, companies can compare the variations in campaign success.
Identify the differences (What needs to change)
As is customary, we want to concentrate our efforts on areas where we can make the most significant changes, so we begin by identifying the most significant gaps. However, there is more to it than that. We cannot afford to serve all of our clients the same way, and not all data is of equal interest or value to us either. Some information is freely available, while other information is not. A record becomes worthless if it does not have certain information.
What is the value of a customer record if it does not contain the client’s name or other means of contacting the customer? Some information is extremely critical but not required, while other information is simply interesting to know or is intended for future use but is not required at this time.
Design the data quality action plan (with responsibilities and dates)
Some of you may have observed that the method taken to project quality management differs slightly from the way we take to overall project management. This is not surprising. There is a quality threshold below which we should never allow the standard to deteriorate. During the period in which the customer’s file is in use, we will develop a regular quality maintenance plan that will be executed on a planned basis. In addition, unique projects frequently have specific quality standards that must be adhered to.
Ad hoc maintenance is frequently associated with a single project and is focused solely on the customers who have been identified as being targeted for the project. However, all database hygiene initiatives are cumulative, meaning that they all contribute to improving the overall database quality. Whatever the plan, including a prior snapshot of the database quality, specifying which tools will be utilised, allocating responsibilities, and determining what results are expected and when they will be achieved.
Execute the action plan
Companies are now ready to put the strategy into action and measure the results to keep track of their progress. To put the strategy into action, they will use a combination of the technologies listed below.
Getting Information Quality
Quality information storage in firm data bases necessitates the measurement of that information’s quality, the identification of quality concerns, and the establishment of mechanisms to correct such issues.
Building in Quality
The most effective method of obtaining high-quality consumer information is to ensure that it is captured accurately in the first place. Those in charge of data gathering, on the other hand, are frequently managed by siloed organisations that do not track data quality performance metrics.
Even if this can be adjusted over time, it is rarely a good location to begin the process. Begin by demonstrating the value of the data; next, by highlighting the costs of cleanliness and the sources of the lowest quality, pursue these sources. For the most part, companies are only concerned with the quality of data when it is being fed into a reference file or into a consolidated data warehouse.
Managing Quality for the Life of the Data
We understand there are a variety of factors that influence the accuracy and completeness of the customers’ data. Recurring problems can be prevented and/or fixed by using specific tools in a proactive manner. Many different tools can also be used to repair unanticipated problems that may arise.
Proactive Maintenance
The handling of return mail, for example, is one of numerous essential data hygiene actions that all businesses should perform on a consistent basis. Even while it costs a bit extra to have undelivered mail returned, saving money by avoiding mailing to the same address repeatedly can save companies a lot more money in the long run. The use of scheduled database audits that assess for quality levels by segment and data tier are also quite beneficial. Because some scheduled data quality initiatives are costly and time-intensive, companies prefer to do them only when they are necessary.
An order may have a positive outcome, which may prompt the development of an ad hoc data quality action plan to address the specific concerns that have been identified. Of course, if any of the ad hoc maintenance plans developed prove to be effective, companies may incorporate them into their scheduled maintenance schedule as a matter of routine.
A portion of these data hygiene operations, but not all of them, can be carried out by computer with the use of specialised gear and software. To give just one example, consider the several software packages that are available for address standardisation and matching.
Aside from that, several data services companies have unique processes that perform the same thing while also having the ability to inject more information into each of their customers’ records. It is possible that part of the work will be done directly on the database itself as well. However, only certified, and trained data administrators should be granted access to directly upgrade the customer reference file; everyone else should be restricted from doing so.
Ad Hoc Maintenance
Maintenance performed on an ad hoc basis is utilised to satisfy the requirements of a single project endeavour or when a data quality issue is discovered unexpectedly. These one-time cleaning initiatives should be included in the regular maintenance schedule if the problem appears to be reoccurring in the future.
