Axify variations are calculated from the highlights of a metric over a given period.

The period filter for the Process and Technical axes delimits the length of the reference periods used.

(Example of period filter for the Process Axis)

Note that the variation is only calculated over periods of **one month, three months, and six months.**

Psst! A highlight displays a number or elements that summarize a metric. It is possible to have one or more highlights for each metric.

## Calculations

The variations are calculated as a **variation rate** over the selected periods.

For example, for a selected period of the last three months, Axify compares the highlight of the last three months with the highlight of the three months before that period. These two values are then used in the calculation of a traditional variation rate:

### variation = ( current_period_highlight - previous_period_highlight ) / previous_period_highlight

## Improvement vs. deterioration

For a given metric, a positive variation (i.e. an increase in the value of a highlight) can represent an improvement or a deterioration of the metric. This qualification of the change is represented by the color of the arrow that accompanies the percentage change.

A green arrow indicates an improvement while a red arrow indicates a deterioration.

## Reading and interpreting the variation indicators

By clicking on the desired variation indicator, a tooltip containing more information on the variation for the measure in question will be displayed. The average variation of the current period will be compared with that of the previous period.

Improvement of the cycle time breakdown

Deterioration of the work in progress

Psst! The explanation of the variation, i.e. the first paragraph before the average of the current period, is adapted according to the selected period. In addition, the variation is calculated by iteration.

## Retroactive data or not?

In the case of the Process and Technical Axes, once your integration has been activated and linked to the project, the data prior to the activation date of the integration will be synchronized. You will therefore be able to obtain retroactive variation indicators, i.e. the variation of your processes will be considered up to 6 months before the activation date of the Axify integration (if your project has been underway since then).

As for the deployment frequency and change time (DORA Metrics), activating a webhook will initiate the data import and variation calculation, but will not import the data retroactively. As a result, you will not be able to analyze your deployment performance for the period prior to the activation of a webhook.

## In the case of a new project

It is possible that a project is too young to calculate a variation over 1 month, 3 months, and 1 year.

For example, if a project has been started only 5 months ago, as the calculation of a variation requires at least one year of data (so two periods of 6 months), Axify will not have enough data to calculate the variation.

In this kind of situation, Axify treats the variation of the different periods as follows:

- 1 month -> We compare highlights today - 1 month with 1 month - 2 months
- 3 months -> compare highlights today - 3 months with 3 months - 6 months
- 6 months -> compare highlights today - 6 months with 6 months - 1 year
- 1 year -> not available as of June 16, 2022

In this way, the variations provide a relevant indicator of a young project's progress as the project matures until Axify has enough data to calculate the variations over 1 month, 3 months, and 6 months.