Detect - Settings

Here is the different settings available in DataMa Detect


DataMa Pivot can be used with a comparison. To activate this mode select a dimension and the element you want to compare. Each graph and the comments will be in a Compare mode reflecting the difference of the elements you are comparing.


Filter any element you want to remove from the data or from the analysis (You can remove the dimension appearing in the analysis by clicking the cross)


In DataMa Detect, you can detect the anomaly of only one KPI or all metric relation.

By default, DataMa selects the first ratio of your metric relation, but you can change that by clicking on the KPI bar in the subheader and ticking the appropriate metric within your metric relation


Aggregation in %

The level of aggregation that the model is using – e.g. if Level of aggregation is set at X%, segment within each dimension that represents less than X% of the Primary Numerator (e.g. Revenues) of the main KPI you’re analyzing will be clustered in one « Other » segment. X is set at 2 by default, but you may want to play with this parameter quite a bit because it can change significantly the calculation of mix effects.

Click on the drop-down arrow to display the settings menu Move the cursor to the right to increase the level of aggregation Click on « Pivot » to get results Segments are now aggregated at the requested level

Analysis Language

For now, DataMa is only available in English and in French. By default Datama is in english.

Colors Palette For Plots

Graphs colors are customizable, in 3 different ways. Grayscale, scattered and Custom. The custom method allows to refine color by color all the graphs available in the solution.

Force Rounding

The rounding setting will bypass the smart rounding available in DataMa which adapts the number of decimals according to each number.

Max Segment Displayed

When looking at the bridge, this is the number of segment displayed for a selected dimension. The minor segments will be displayed in a bucket “other”

Maximum Depth For Continuous Grouping

For continuous elements, multiple clusters are created aggregating elements having a similar average KPI. The level of aggregation of the clustering is set by this parameter.

Analyse All Metrics Relation Steps

DataMa Detect can anlyse one step or all the steps of your funnel. When all step are selected, you will be able to see in a dropdown menu every step with an anomaly.

Inspect Numerator only

When you’re analysig all metrics relation steps, you can choose to analyse only the numerator. This is useful when your market equation is not important or not relevant and you prefer to analyse independently all the Numerator.

Analyse Latest Point Only

You can choose to display the anomalies of each date or only for the latest point in the data. On a day over day anomaly detection we recommand to activate this parameter, to avoid having all the anomalies raised.

Detection Method

When “Analyse latest point only” is activated you can choose the standard “Moving Window” or the “Forecast” which will forecast the most likely confidence interval.

Number of points for average window

This allows you to define how many point in your historical data the algoritm need to take into account to forecast the confidence interval.

Confidence Interval

A confidence interval is a range of values that is likely to contain an unknown population parameter. If you draw a random sample many times, a certain percentage of the confidence intervals will contain the population mean. This percentage is the confidence level.

By changing the confidence interval you can modify the precision expected from the significance.

Analysis Depth

This is the level of analysis in the explanation part of the solution. How many detail will be given in the comment.

Include 0 For Y Axis

By default active, this makes the 0 line appear on the waterfall When deactivated DataMa will adapt the Axis to your dataset. This help to better display small variations.