Why Data Can Make or Break Your Sales Performance Management


Understanding data and data quality is one of the most critical, yet most overlooked tasks when preparing to implement a new ​ Sales Performance Management system.

Indeed, data quality is important in any information system. As mentioned by Thomas C. Redman in a Harvard Business Review article, “Data quality issues come in many forms — from not having the data you really need, to data that are easy to misinterpret, to data that simply can’t be trusted. Worst are the data quality problems that you don’t even know you have. Unfortunately, they are all too common.”

In the context of sales compensation, inaccurate or late data can mean paying too much (or too little), paying late, and losing the trust of your sales force. So prior to implementing a new system, here are a few questions to ask yourself to assess the readiness of your data.


What data is required?

​A large variety of data may be needed to support your Sales Performance Management system. This includes both transactional and reference data.

Transactional data refers to a business transaction such as a sale or event, which usually results in compensation being paid. Examples of transaction data are the sale of a handset by a mobile phone company, or the activation of the handset.

Reference data refers to any data needed in order to support your compensation plans and calculate, pay, and report compensation. Examples of reference data include employee information, product SKU numbers and pricing, and currency data.

It is also important to consider what data is NOT required. Data should only be loaded into an SPM system if it is required to support the calculation of the sales compensation plans or if it is required to provide additional analytics from the SPM system. For example, transaction records from a source system might contain data about the client that purchased the product. Unless the client data can affect compensation then this data is not required by SPM and should not be included in a data fed into the system. It is also common to only load transaction data that has been carried out by someone who receives variable compensation through the SPM system.

Where is the data available from?

Data can either come from an upstream system such as an HR system (employee and sales hierarchy data), or Point of Sale system (transactional data) or be manually generated. Data from upstream systems is more likely to be accurate and will reduce the risk of user error. The owners of any upstream systems that are required to provide data feeds into SPM must be consulted early on to validate what data is available and when.

If data has to be provided manually because it is not available from another system then it must be asked who is responsible for generating the data, what checks and measures need to be put in place to validate the data before or after it is loaded into the SPM system.

In the case of manually generated data it might need to go through an approval process before being loaded into the SPM system. This may be true for all data or for transactions over a certain value. Data may also be verified against existing reference data in the SPM system to ensure that mis-keyed data is not loaded into the system.

When is data provided?

Timeliness of data is important. For example, if data is only available from upstream systems on a monthly basis then you will not be able to support weekly compensation plans. It should be determined, as accurately as possible, when data can be provided to the sales compensation system as this will affect processing windows. Ultimately, this will determine when the sales force can see their latest data available on reports.

When data is available is very specific to each customer and varies by industry. A shop selling mobile phones, for example, may have transaction data collected throughout the day whereas a bank selling mortgages may only collect transactional data every week or even less frequently.

How accurate/complete is the data?

In an ideal world, all data loaded into ICM would be 100% accurate and complete. In reality, however, it must be accepted that for a variety of reasons, data will sometimes be inaccurate or incomplete. An example of this is employee statuses pertaining to sick leave. It is very common for SPM systems to not receive data that a salesperson was on a leave of absence until sometime after the event. This is particularly true for any data that is manually entered.


By understanding the possible ‘issues’, it can be planned for by performing extra validation or authorization of data before it is consumed by an SPM system. In the cases of missing data, which is needed to support a compensation plan but which is not available from a source system, it may be necessary to preprocess the data that is available using business rules to generate the missing data. Common scenarios, such as the ability to change the case of text data, concatenating fields, and removing leading zeroes from numbers, may be included out of the box in some SPM software. It may also be possible to build customer specific functions to manipulate data using the configuration tools built into the software’s user interface.

Exceptions: What happens when data changes or arrives late?

It is almost certain with data that there will be exceptions to the rules. Although the nature of the exceptions may not be known in advance, you must plan for them. For example, if transactions are received late how should they be processed? What if an employee was on sick leave last month but the SPM team were not notified until now? Understanding how data can change over time and what special cases the SPM system must be able to process will reduce your risks.

In summary, it is essential that the data required to support the business compensation plans is understood as fully and as early as possible in the project. This should be supported by obtaining accurate and complete data to be used during the implementation of the system.

Learn how NICE handles sales compensation data.



Share this:
Twitter LinkedIn Facebook Email