Do Data Better So It Does Better Business for You

By April 20, 2020 May 25th, 2021 Blogs

Like me, you are probably surrounded by data charts and reports using tools like colorful sales dashboards and sophisticated Excel reporting visualizations feeding decision making. But how do you know if report numbers presented are complete, accurate, and trustworthy? Do you ever wonder where the figures came from? You may periodically ask, “How can my gross profit be so far off?

Depending on what your company sells, for example, lots of SKUs through various channel partners, the problem of trustworthy data may be a significant challenge. A simple out-of-date cost of goods sold measure may skew profitability and return on investment KPIs in a very big way and may go undetected for months. This is common in reporting which is heavily manual in nature or primarily handled via outdated third-party solutions.

Where do data problems typically occur?

The response to this question can be complex; hence, only a few of the likely areas are cited here. Typically, a formal review of your specific data workflow processes is required to truly identify weak links:

  • Data is Incomplete. All too often, sales figures, shipment information, inventory, and other relevant data. may not be entered in complete detail at the time sales reporting is being performed. Channel partner store sales may not be reported concurrently making the calculated ROI of 28% today turn into 22% in a couple of weeks when all the store data is received.
  • Data is Out of Sync. Most of the time, enterprise reporting solutions rely on input data from several systems, and if updates from these systems are not accurately applied, report data will not be synchronized. This situation can generate errors in the data and, if some input is lost, the errors may never be found.
  • Non-conformant data causing issues. Inputs to enterprise reporting may contain incorrect data (bad COGS, incorrect discount price) which get propagated into the formulae and output dashboards. Detection of non-conformant data through business rule checks may be inadequate to detect all conditions and properly dispose of the bad data.

Finding the time to identify sources of data problems and build the counter-measures to ensure data completeness and accuracy is hard. This is fundamentally why these types of data problems linger in outdated reporting solutions.

Your organization needs to do better to support an elevated level of business growth and profitability or else data problems can pose a risk to future decision making.

So what can I do about these problems?

The triage of bad data in analytics is necessary to address the areas of great business impact. Tracking of data problems in a challenge for which a tracking solution is absolutely necessary so that critical data issues are not lost and get prioritized.

The following activities can help to begin the process of bad data mitigation:

  • Formalize a Business Data Dictionary. Each organization needs a glossary of global financial business terms and formulas supporting calculations of key KPIs. Any regional/market dependencies should be noted and examples of calculations provided.
  • Publish a formal schedule of updates from input data sources and reconcile data. Business users should know when reports and dashboards will be updated with the most recent data. Processes to validate and reconcile incoming data should be developed. The degree of reconciliation should depend on the importance of data to reporting.
  • Reports should be marked with data status indicators. If live dashboards are reporting weekly/daily sales performance, visual indicators must be provided to reveal if data is complete for a particular sales channel. This will alert dashboard users about potentially incomplete data.
  • Exception data clearly identified. Non-conformant data should be identified through business rule validation and made available for review and overrides. Trapping bad data before it enters the reporting system and corrupts output KPIs is essential. Data operations staff must be provided with tools for exception inspection and override.

These basic steps are a part of a larger data quality program that should be established across your Information and Analytics department. Driving better corporate performance is possible with an orchestrated initiative to enhance the quality of your strategic reporting systems and replacing outdated solutions.

Saama Analytics uses best practices with its clients to overcome data challenges like these and deliver solutions that enable optimal business performance. The Saama Analytics’ Spend Optimization solution is an example of advanced sales analytics that accepts a wide variety of data inputs and provides data quality measures to ensure accurate reporting. For more information, see

George Shemas

George Shemas

Principal at Saama Analytics Client Care organization overseeing its Consumer Goods practice with decades of experience building world-class analytics