The dashboard is a functional journey analysis tool. The main objective is to provide a higher-level analysis than the results displayed on the Greenspector App Workshop, with a notion of journey. We detail here :

Calculation method

Global Ecoscore

The Ecoscore is a score out of 100 with a dual purpose:

...

• 🟡 Gold level - 100 à 90

• ​⚪ Silver level - 90 à 70

• 🟠 Bronze level - 70 à 50

Critical and non critical steps

Critical stages are entry points to functional domains with a theoretically higher number of visits than other steps. This makes it possible to assess whether or not the quality of the path is strongly impacted by bad critical steps.

...

• 0.7 for the score of critical steps

• 0.3 for the score of non-critical steps

Metric calculation

Example of performance calculation:

...

This gives us an indicator that is representative of the quality of a functional journey, regardless of its length. It does not depend on the estimated environmental impact.

...

Dashboard sections

Header

...

List of fields and their correspondences in the definition file :

Synthesis

This section gives an overview of the results. It gives a good idea of the quality of a functional journey, without necessarily going into detail.

...

1. Total number of steps measured without reference. Check that this corresponds to the number of steps defined in the definition.yml.

2. Ecoscore previously described

3. Main performance, data and energy metrics. We find :

   • The total value of the metric for the journey :

     • ⏲️ Time in seconds for performance, action and loading steps only.

     • 📡 Consumption in MB for data

     • 🔋 Consumption in mAh for energy

     • Unlike the scores, these values depend on the length of the course and are given as an indication.

   • The evolution when comparing several versions, in the unity of the metric :

     • 🟢 Decrease above 10%

     • 🟡 Stable evolution

     • 🔴 Increae above 10%

   • Overall metric score for the service tested

4. Details of critical and non-critical step scores.

Notes by step

This graph represents the scores for each step along the journey. Line-by-line analysis is not necessary, but the graph gives a good overview of the trend. We will concentrate instead on analyzing the last steps, which are the worst and those to be treated in priority.

...

Metric detail

...

1. Metric of the section

2. Analysis of all types of steps. It allows you to identify common tendencies between different types of step.

3. Main analysis :

   • Performance : load steps

   • Data : load steps

   • Energy : load, action, scroll steps

4. Secondary analysis :

   • Performance : action, scroll steps

   • Data : pause, action, scroll steps

   • Energy : pause steps

5. Analysis by functional domains in definition.yml, parameter domain. It highlights the functional domains with the strongest impact for each metric.

...

Specificity of energy

In order to harmonize consumption between terminals, the threshold used is a multiplier of the reference consumption of the phone in inactivity. The energy rating of the steps is based on the ratio of the step's energy discharge speed to the reference.

...

This graph can be used to highlight steps that are abnormally long, or steps that consume abnormally high amounts of energy, even though they are short.

Environmental impact

You can directly visualize the environmental impact of a functional journey by completing the section environmentalinput in the file definition.yml.

...

Dashboard analysis process

Constructing a global statement

A quick overview of each section reveals any apparent problems with the service being measured. An overall assessment can be deduced from this:

• Slow but energy-efficient service

• Fast but energy-hungry service

• Energy-efficient but high data consumption

• etc.

Identify problems

Analysis of the dashboard should help identify optimization levers. Two situations are possible:

• Most of the steps are bad on one metric. Analysis will be simplified with a lot of problems to noticeavec beaucoup de problématiques à relever.

• Some steps behave far from the average. We need to be sure that this is not a measurement error. We then need to go into a more in-depth analysis and focus on these steps.

Start analysis of the dashboard in this order:

1. Performance

2. Data

3. Energy

This makes it possible to construct an argument such as :

• Performance analysis shows that step A is time-consuming

• Data analysis associates step A with high data consumption

• Energy analysis links this data consumption to high energy consumption

What the dashboard doesn't show 😨

...