Skip to main content

Symon.AI help center

Manual Adjustments Outliers

Abstract

Find outliers in manual adjustments which can be a source of errors in your sales compensation processes.

Manual adjustments can be a significant source of errors in your sales compensation processes. This blueprint finds outliers in those manual adjustments. It also identifies adjuster bias to help detect if any sellers receive preferential treatment.

Legend

The tools in this Blueprint are color-coded based on their function:

  • Yellow: These tools contain the input data or results that are ready for export. These are what you’ll be mapping to in Incentives for the Symon.AI Calc object. The Adjustments Data (Data) tool contains the manual adjustments data you want to find outliers in.

  • Pink: These tools contain the core of the outlier analysis that this blueprint provides.

  • Red: These tools contain the business rules using the results of the outlier analysis.

  • Blue: These tools manipulate and transform data for visualizations.

  • Lime Green: These are the output tools which contain results that are ready to be exported.

Pipe inputs

This blueprint uses the Adapt (Adapt) tool to map the columns in your data. This blueprint expects each row of data to correspond to a manual adjustment with these required columns:

  1. PayeeID:The unique identifier for a seller.

  2. AdjusterID: The unique identifier for the administrator who entered or approved the manual adjustment.

  3. ManualPayments: The manually-adjusted amount granted to the seller.

You can look at the sample data to see how the tool maps columns. Whatever column names you have, you can use the Adapt (Adapt) tool to change them to names the pipe recognizes. In the sample, the remapping looks like this:

  1. PayeeID to PayeeID (no change needed)

  2. Adjuster to AdjusterID

  3. ManualPay to ManualPayments

How to interpret the results

The most important export tool is the All Adjustments with Flagged Outliers (Export) tool. This tool contains all of the manual adjustment rows. It also includes the Outlier_Type column. This column tells you if Symon.AI flagged the adjustment as an outlier. It also tells you what type of outlier it is. Any rows with a value of Passed means that it was not flagged for review. Any other value in this column indicates you should review it and provides a reason why.

Visualizations

This blueprint comes with visualizations and dashboards, no configuration required. These visualization can help you see where your adjustment outliers originate from.

Adjustments by Adjuster: This Sankey diagram is one of the visualizations included in the blueprint. It shows the breakdown of adjustments made by the adjusters, if they were flagged as outliers, and who is getting the adjustment. In this example, we can see that the adjuster with ID ADJ-200 has a large number of adjustments made to the account executive AE001. The reason given for the possible outlier is adjuster bias. With this information, you can investigate the adjustments further. This could be a sign of problems in the compensation plan affecting AE001 or it could be a sign of unfair preferential treatment.

How it works

This Blueprint focuses on two approaches to identify outliers.

  • The first approach analyzes the adjustment amounts and finds outliers across all adjustments.

  • The second approach aggregates all adjustments made between adjusters and sellers in the Aggregate tool. It then identifies any outliers found in that data.

Tools labeled as pink are where the blueprint analyzes the outliers. If you look at these tools, you'll find three:

  1. Adjustment Amount Outliers - Z score (Outlier)

  2. Adjustment Amount Outliers - kNN (Outlier)

  3. Adjustment Amount Outliers - PCA (Outlier)

Each of these tools find outliers in data sets with different distributions. For example, Adjustment Amount Outliers - Z Score (Outlier) is great at finding outliers in normally-distributed data. It doesn't perform as well when data is randomly distributed. That’s where the other tools come in, each tool can find outliers with different types of data distribution.

If a row comes back as an outlier, the business rules encoded in Outlier Type (Case) combine the outlier results from all of the analysis methods into a single value.