As discussed in our previous posts, all your sales and operational data does not provide the same level of value to your organization. So then why do you think you should manage all data to the same level of quality? Quality is relative to the value the information provides or supports as measured by its contribution to your business objectives.
The goal is not perfection – because there is a cost to perfection that will not necessarily lead to increased value. A better goal might be... “imperfect – but with intentional and consistent quality”.
Here are some tips to help you develop and manage a balanced data quality program.
“Value Stream” Your Data
The concept here is that if you can’t map (clearly identify the affect of) how the data contributes to a meaningful and measureable outcome for your organization or department, then it is simply an interesting piece of information and not worthy of process control. Note that this does not necessarily mean that the data should not be maintained at all – but it does beg the question.
Begin by outlining the key measurements that drive your organizational or departmental success. These are the outcomes (think KPIs) that must be met in each of your organizational performance areas such as compliance, financial, operations, customer service or reputation management.
Next, go down a level and identify the measurable outcomes that drive success in each of your key measurements for each organizational performance area. For instance, one such measurement in the financial domain may be total sales revenue generated in a period.
Then map the data that contributes to each of those measurements. Be specific and identify the data element’s contribution level by categorizing it as:
- Direct – has a direct relationship to the final measurement. For instance, the actual GL revenue generated from a sales transaction. This then ties back to a key measurement of profitability.
- Supporting – is not directly attributed to the measurement, but supports the narrative for the measurement. For instance, the level of discount applied to the list price to arrive at the net revenue amount.
- Leading – a direct measurement to the future of the primary area of focus. For instance, the estimated net GL revenue on open sales opportunities.
Balance is Key
Once you map each of the organizational performance areas, you should validate them based on the tiers you outlined based on our previous Blog post. Then you can more easily identify each of the organizational performance areas. This helps you understand WHY data is important and is valuable in supporting process and data quality communications with your team. This also informs how you must balance your organizational performance areas and therefore your data management and quality programs.
Lastly be sure to look at the data that you maintain that IS NOT on the list created through this analysis. Ask yourself why you maintain it. If you can’t identify how the data provides current or future value, is it worth having? In a world that leads you to believe that more data is always good – it can be argued that without intention and purpose it is just cost and not value.
Design Quality Goals Based on Value
Now you have the information to define data quality goals that are meaningful to your organization. Using data source and business process analysis techniques, you can build processes that increase the value of the right data while not spending time and money on information that does not lead to results that are important to you. You now can build an intentional and consistent quality management program for your data. This is the subject of our next post – in the meantime, click here to read the rest of this blog series.
Want help getting started? Get the Data Stewardship Checklist