[Metrics] Views and open rate per time span for each contributor / bubble / website

[MaartenLMEM]

Need To better understand impact of contribution work we need some basic consultation metrics.

I. First one is open rate, which means : when a bubble is diplayed, how many time it's open. We need an overview :

• Per month is usefull for most of cases.
• Per week is usefull to assess our actions on very popular bubbles. For those, it s long to wait a month to know if my work I made is usefull or not.

And :

1. for each bubble (also named notice)
2. for each contributor : to know which is the best working, which one we should extend, which one improve
3. for each website : to better target our contributions

II. Second metric we need is : views Today, in backoffice we know how many views we have per bubble. But we cannot make comparison over time. For example compare views betwenn last 15 days and the 15 days before.

Features

1. A CSV extract of Backoffice on last 3 months would be enough to buid a big part of these metric. It contains :

   • Date : year-number/month-number/dau-number
   • bubble number
   • contributor name (or number)
   • websites targeted (if there is a group, just list all websites it contains)
   • number of display
   • number of detail views
   • number of click on links

2. CSV extract of Matomo, on last 10 weeks, on last month and on last 3 months containing :

   • contributor name (or number)
   • bubble number
   • website domain
   • number of display (of the bubble on this domain)
   • number of detail views
   • number of click on links

Target user

• Contribution team who create an improve bubbles, and those who make large improvment actions
• Finance team to share use KPI
• Communication team to tell what can expect a contributor when he comes on DisMoi

[MaartenLMEM]

to @J-oh-n for information to @zhinu @christpet for a first feedback about those extracts complexity and to discuss the best way to do it each month (or better each week) fornext 6 months.

[christpet]

@julienadamcom you've been doing some work on the backoffice data so you might have a quick answer for how complex this is, otherwise @lutangar or @JalilArfaoui ? Whoever sees this first, please give us a quick estimate of size: XS - S - M - L - XL

[zhinu]

On my side, I know how to extract views per bubbles (with their id), and probably clicks as well. However, I don't know where to extract the other information pertaining to the bubble (i.e. date of creation, contributor id etc...)

[julienadamcom]

On my side, I know how to extract views per bubbles (with their id), and probably clicks as well. However, I don't know where to extract the other information pertaining to the bubble (i.e. date of creation, contributor id etc...)

let's talk about it, what you have is what I miss!

[MaartenLMEM]

@julienadamcom Here are my notes after our meeting with @zhinu :

• Couverture d'expérience web : plutôt BO, mais cohérence données pourra être vérifié avec Matomo
• Taux de consultation : plutôt BO, semble t il pas possible par Matomo - car l'évènement consultation détail n'est pas isolé.

Donc le sujet est plutôt entre tes mains maintenant, en sachant que l'année dernière @JalilArfaoui m'avait fait tout un extract que j'avais exploité ici.

For now, i would be very happy with data about past two months.

By the way, here are more precise definitions of these contribution metrics :

• Couverture d'expérience web : Nbre bulles vues / utilisateur / mois
• Taux de consultation : % consultation détail / bulle vue

[MaartenLMEM]

@zhinu peux tu faire un point sur l'avancement à ce sujet à partir de ce que tu m'a montré :

• ce qui est dispo par rapport au besoin exprimé et où
• ce qui reste à faire

Si c'est ok pour toi on met cette carte sur le Dashboard Analytics uniquement et je l'enlève de BO.

[MaartenLMEM]

Actuellement problèmes restants :

• différence entre entre data de matomo et BO see #363

Ce n'est pas bloquant. Next step : Je vérifie que ça répond bien au besoin exprimé.