Expert Insight: Managing Supply Chain Management Performance
  By Kate Vitasek  
  July 9, 2008  

How Good Is Your Supply Chain Data Quality? (Part 3)


Data Cycle Counts: You Track Inventory Accuracy - Why not Data Accuracy?

Vitasek Says:
Cycle counting is more than tracking inventory accuracy; the key purpose of cycle counting is to identify items in error, thus triggering research, identification, and elimination of the cause of the errors.


Many companies claim that the reason that they do not measure data quality is that it is too hard to do, or that looking at all that data takes resources they do not have or that the volume of data is so large that it is an overwhelming task.  A few companies have found a creative way to assess data quality, implementing a Data Cycle Count program where they “cycle” through the data in their system and “count” the number of data errors found.  By looking at a sample on a regular basis the task is more manageable. 

Companies have been using cycle counts to assess the accuracy of one of their most important assets, inventory, for a very long time.  Accuracy of the on hand quantity is important to the success of the company. But cycle counting is more than tracking inventory accuracy; the key purpose of cycle counting is to identify items in error, thus triggering research, identification, and elimination of the cause of the errors.  It is this process of continuous improvement that yields benefits. 

How do you apply cycle counting to data?  Data can be viewed as an asset too, and like inventory assessing, its accuracy is important to understanding if the processes in place are sufficient to control the input and maintenance of data.  As in inventory cycle counting, it is the process of error identification and elimination of the cause of the error that yields benefits. 

The data cycle count process is simple; you select a data record to audit, you compare it to the field requirements for that data record and then you report out the rate of the errors, perform root cause analysis tracking the errors by type.  Then attack the largest error group as an improvement project to fix the source of the error.  That is the basic process. 

As with inventory cycle counting, deciding what to count and how often to count can vary by company.  The same is true for data.  Data can be grouped by type, importance, frequency of change, or how prone it is to entry error.  However you choose to classify the data assigning an ABC classification will help you manage the selection of the data records to count. 

What are the benefits of data cycle counts?

  • Data cycle counts are a quantitative measurement of data quality, no more gut feel measurements that are prone to second guessing.
  • Data cycle counts measure accuracy over time, tracking how well improvement efforts are working.
  • Cycle counting data is a sampling process and is easier to execute than a full data review.  The sample accuracy rate can be extended to approximate the error rate for a group of data.
  • Cycle counting the data encourages an environment of problem solving and continuous improvement through ongoing corrective action.

Data cycle counting is a simple process that any company can implement and get tangible benefits from.  Data is a valuable asset and it is time for companies to make the effort to measure data quality.  By borrowing from measurement methodologies used in other parts of the business, companies can develop formal processes to measure and track data quality.

Accurate information and the flow of data is critical to today’s extended supply chains; data drives the choices companies make, whether tactical or strategic.  It is data that is used in the measurement of processes and outcomes, so if it is not accurate, your performance measure will not be accurate. Yet, few companies measure the quality of their data or appreciate how poor data quality impacts their performance.  Include data quality in your performance metrics program; it is easy to do using methods you already have in place and it will drive significant and long-lasting improvement!

See Part 1 and Part 2 of this Data Quality series. More columns soon.

Agree or disgree with our expert's perspective? What would you add? Let us know your thoughts for publication in the SCDigest newsletter Feedback section, and on the web site. Upon request, comments will be posted with the respondents name or company withheld.

  Send an Email