Quicksight Analytics Dashboard
Quicksight is an AWS product that Viafoura leverages for data visualization for its customers. It connects directly to Viafoura’s data lake which updates every day (you will see the data until yesterday). You can export and filter the timeframes for the data you are looking for.
Conversations
Shows unique daily comments (comments and replies) and commenters
This graph looks only at comment posting actions (not deletions)
Comments are counted on the day they are posted, not moderated.
Shows unique daily likes and dislikes
This graph counts likes/dislikes actions i.e. if I like a comment and then dislike instead both are counted
The % of your comments are in reply to top level comments
This is a good indication of the health of your community; The industry average is about 50% with most sports publishers seeing a higher number
Daily comments by domain is used to compare engagement across all of your sites
You can see both the busiest site by commenting volumes but also the busiest days
Shows the busiest period across your site for commenting
This is useful for planning live events and knowing when your community is most active
Remember all times are in UTC
Shows daily conversations follows
This graph counts follows actions and we are not checking for "unfollow" events
The table shows the URLs with the highest follow counts
Users who follow conversations get a tray notification each time a new comment is posted
How many clicks on "load more comments" per day
This can depend on how many comments you have on initial load
The graph is broken down by segment type
Viafoura Segments
Anonymous - Non loyal fly-by audience
Comment Readers - Also known as Active Engaged, they are now starting to spend more time in the comment section and becoming more loyal to your site
Registered - Your known audience, these users have created an account with you
UGC Contributor - A registered user who is contributing to the community
Chats
Shows unique daily messages posted in Chats
Counts the posting actions and not deleted actions
New chat containers created per day
New chat containers created per month
% of comments posted that are replies vs parent comments
Reply to comment ratio is usually lower in Chats than Conversations due to the fast paced nature
Blogs
Comments includes both top level journalists post and comments from users
Comments in Liveblogs can be enabled using the front end settings panel
New live blog containers created per day
New liveblog containers created per month
% of comments posted in Liveblogs that are parent comments vs replies
User Interactions
When a user creates an account on a client site the account information us shared/sent to Viafoura via API or front-end code and creates a brand new record in Viafoura's back end system
In some cases not all registrations are passed to Viafoura, depending on your set up registrations may only be passed to Viafoura if a user engages with any of the Viafoura features on-site
Sometimes when clients migrate their auth system and decide to ask users to effectively re-register their account (often to reset a password in the new auth system) we will see an influx of "new" registrations. This can happen during onboarding and depends on how users are transitioned to Viafoura
How many comments each commenter is creating on a daily basis
Engagement actions (comments, replies, likes and dislikes) created by UGC Contributors
Articles
Average comments per article across articles with at least one comment
Articles bucketed by Comment count on comment container creation day (article publish day)
Shows top articles for engagement actions by URL (not container)
You can sort by column, by clicking on the column names and selecting sort ascending or descending (ticket submitted to fix this)
Articles that have been set to pre-moderation
Articles may show up multiple times as we check the status every day
Moderation
All moderation decisions broken down by auto-mod or manual
Total moderation decisions will be higher than total comments as comments can be moderated multiple times.
Examples: If a comment has been edited, it will pass through moderation each time. If a comment has been approved by the auto-mod and then flagged by a user it will then be human moderated.
All moderation decisions for time period selected
Auto-moderation decisions of comments
Auto-moderation decisions
Approved: The algorithm has approved the comment and it goes live in real-time
Deferred: The auto-mod sends the comment for human review
Rejected: The comment is auto-rejected and does not go live
All auto-moderation decisions for time period selected
Manual (Human-mod) decisions
This will include all deferred, flagged and NUQ comments
Manual moderation decisions for time period selected
Comment moderation decisions
A comment often has more than 1 moderation action associated with it
The number of articles with pre-moderation enabled
Based on the day the container is created (article published)
Unique comments outcomes
Despite how many mod actions each comment only counts as 1
Comments approved or rejected by time period selected
Reported comments outcomes
Data is based on the date of moderation and not the reported date
Each reported comment should only be reviewed once, so even if a comment is flagged 10 times, this will count 1 moderation action
Flagged moderations for the selected period
Unique comments flagged per day
If a comment is flagged on Wednesday and Friday it will show up twice
Total Comments Flagged and % of Total Comments for selected period
User account moderations covers actions taken on a user and not their comments
Avatar moderations, Username moderations & Userbans
Avatar moderation outcomes
Please note this feature is not available to all clients
Auto-moderation capacity MTD based on monthly contract volumes
Create an alert to ensure you're aware if trending over capacity
Human moderation capacity MTD based on monthly contract volumes
Trending auto-moderation capacity estimate based on daily volumes MTD
Trending human-moderation capacity estimate based on daily volumes MTD
Users
This tab allows you to select the amount of Top Users you would like to show from 1-200
This is sorted Alphabetically by username
You can sort by engagement column by clicking on it
Users in this graph are moderators who are pinning and picking comments
????
The % of users who register and post their first comment
does this include likes and dislikes?
Attention
Total Direct Page views generated by users clicking on Viafoura CTA's
Pageviews generated by Viafoura broken down by source
| Source | Description |
|---|---|
| Conversations Top Pages | Active conversations below the comment section |
| Tray Notifications | PV's generated by users clicking on tray notifications e.g. replies, likes and dislikes |
| Tray Top Pages | Active conversations in the tray |
| Comment Share | Users sharing individual comments |
Total time spent in Viafoura this includes Conversations, Blogs, Chats and Ratings & Reviews
This is cookie UID based tracking
At least 50% of the Viafoura element must be in the viewport and the tab or browser must not be in the background
Time spent by users on pages that have the Viafoura js enabled
Cookie UID based tracking
The browser/tab must not be in the background
Time spent in the Comment Section
Time spent in Live Chats
Time spent in Liveblogs
Time spent in Ratings and Reviews
Comment readers with >3 seconds with the comment section in viewport per day
Daily Unique Visitors broken down by Segment Type
Monthly UVs by segment type
Daily UVs by user segment type (%)
Monthly UVs by segment type (%)
The ratio of Time on Page vs Time in Viafoura Widgets
Bell Performance
Monthly bell CTR based on loads vs clicks for all users
Monthly bell CTR based on loads vs clicks for logged in users
Daily bell CTR based on loads vs clicks for all users
Daily bell CTR based on loads vs clicks for logged in users
Monthly bell CTR for all sites under the client for all users
The bell performance is broken down by bell type
Daily bell CTR for all sites under the client for all users
The bell performance is broken down by bell type
Monthly bell CTR for all sites under the client for all users
The bell performance is broken down by bell type
Daily bell CTR for all sites under the client for logged in users
The bell performance is broken down by bell type
Broadcast Notifications
Open rate of broadcast notifications received in the tray
User Interactions with all broadcast notificatons
Notified events are notifications being sent to the tray
Every interaction sends an event e.g. 5 likes on a comment sends 5 notifications
Users receiving events
There will always be more events than users, a user will appear once per day regardless of how many notifications they have received
A "seen" event is counted when a user opens the tray and sees the notification
Users seeing the notifications in the tray
User is only counted once daily, even if they have multiple notifications
When the users sees the content they are notified about
This number can be high as users may have already seen the content they are notified about in the comment section before going to the tray (this counts as 2 read events)
The users who have seen the content in the notifications
Number is usually 100%+ because users often see the event they are being notified about before going to the notification
e.g. if a user receives a reply to their comment, they may still be lingering on the comment section and this will count as a read event
E-mail Notifications
Data shows the date that the email was sent and the date that the email was clicked
Number of emails clicked/Number of e-mails sent
How many clicks per user on e-mail notifications
Topic Follows
Counting subscribed and unsubscribed events
Topics that currently have the most subscribed users
Sharebars
Number of shares per day using Viafoura's sharebar
Most shared articles by share channel
Most shared channel
Calculated by using URL parameters
We only track the 1st PV after the user has visited the site
Capturing registrations that happen after a user has interacted with the Sharebar
Engagement Starter
Snapshot of usage, performance and positioning on page
Avg Scroll depth: Every event sends the page/scroll depth of the event
Showing how often the ES was in view and how often it was interacted with, across all Engagement starter views (Default, Featured comment, Poll)
How often the loaded ES was in view broken down by Type
How often was the Engagement Starter interacted with
Polls have the highest engagement rate
How often is the Engagement Starter being added to the pages and what type of Engagement Starter is being used.
Polls
Daily votes casted in polls
Polls created by day
Poll creators who have received the most votes in their polls
Number of individual polls created
Top performing polls and results snapshot
Polls with potential vote stuffing
For polls we currently count the number of voting events and unique votes we count using Cookie IDs - discrepancies can highlight any potential vote stuffing
Polls created and their votes over a set period of time
Use this to analyze what works well - sports, politics, entertainment
Page Performance
Ensure you remove the http:// from the URL when adding the filters
Comments, PVs and Visits for the specific article or campaign
Registrations attributed to the URL and engagement metrics for the specific page
Data for campaigns
Data for URLs
the URL you would like to compare it to
SDK
Ratings and Reviews
Viafoura's Ratings and Reviews feature
Comments posted in Viafoura ratings and reviews
Number of new Ratings and Reviews containers created
Access to Quicksight
- Reach out to your dedicated CSM or [email protected] if you require access to Quicksight
- Access can take ~1 week to be granted
- When logging in ensure the Account Name is viafoura-production the User Name will be your email address
Important Notes
- All times are in UTC
- All data can be exported using the kebab menu in the top right of each graph
- Most graphs have data from 6 months - 12 months to ensure long-term access, we recommend downloading any data you need for perpetuity
- We do offer raw data ingest which will allow you to bring this data internally and integrate it into your own analytics/visualization tools
- Some graphs will say "No data" this means you are not currently using this feature, if you think you're receiving this message in error please reach out to your dedicated CSM
Updated 4 days ago