Difference between revisions of "May 15, 2024 - Updates, priorities of the week"

From wiki
Jump to navigation Jump to search
(Created page with "Back to GC Web Priorities Meetings - Réunions des Priorités Web du GC ==Meeting information== ==Agenda== ===Updates=== '''PCO''' *Congratulations to all for the work on...")
 
Line 3: Line 3:
 
==Agenda==
 
==Agenda==
 
===Updates===
 
===Updates===
'''PCO'''
+
'''DTO'''
*Congratulations to all for the work on the budget
+
 
*Budget bleeds into ad approval budgeting
+
* DTO and CDS have published some blogs:
*We’re expecting ad campaigns coming soon, also with tracking capabilities such as Adobe campaign
+
** [https://blog.canada.ca/2024/05/13/flexible-navigation-page-design-for-emergencies.html Flexible navigation page]
===Roundtable===
+
** [https://digital.canada.ca/2024/05/15/a-shared-focus-working-together-to-improve-online-services/ A shared focus: Working together to improve online services]
'''Principal Publisher'''
+
 
*11 news items for Budget put through AEM, and 6 non-news items
+
'''PCO/Public Safety'''
*The visuals looked great, everyone was happy
 
*GC search:
 
**Our contract with the current search provider is ending this year
 
**We have found a new provider for the search engine, a Canadian company named Coveo
 
**We are getting ready to reach out to teams that are involved with search
 
**We will need to work with teams to refine and test the search
 
'''Question:''' Did the search email go out to all AEM publishers?
 
  
'''Answer:''' Yes it did.
+
* Departments are collaborating to provide information on wildfires
 +
* Public Safety (PJE Theme) - Monitoring wildfire events in Western Canada. Preparing for eventual requests for assistance from provinces. Evergreen wildfires event page is being reviewed and updated by partner departments: https://www.canada.ca/en/public-safety-canada/campaigns/wildfires.html
  
'''Question:''' Is the search assessment tool that DTO is working on being incorporated into the internal search tool?
+
[[/docs.google.com/presentation/d/1QOW9sA9UiF6n7 Sla2gwC-IQiMTbduEjPHGhSP5KT3s/edit#slide%3Did.g2afcaefb9e7%200%2029|Presentation - ECCC - Jade Bourdages - Climate action awareness advertising campaign 2023-24]]
  
'''Answer''': Yes, DTO will work with PP to refine the search with the Search Assessment Tool.
+
* November 2023 campaign was launched
 +
* Goal was to show what Canadians can do for climate change
 +
* Data was very valuable in making adjustments during campaign
 +
* Two focuses
 +
** What individuals can do
 +
** How the government can help
 +
* What visitors were interested in - how government can help
 +
** Loans
 +
** Incentives
 +
** Grants
 +
* 4 out of 5 top clicked links, in the “what government can do” section
 +
* We also used the Page Feedback Tool
 +
** Some people were insulted by emphasis on individual action, on such a big problem
 +
** 80% of feedback said that government was most responsible to solve problem
 +
** People want government to be part of the solution
 +
** Global was the 3rd most used term in feedback
 +
* People expect that the GC website would focus on government action, rather than individual action
 +
* A/B testing, we tested what was displayed first “We’re here to help” first, and the “Individual” section
 +
* “We’re here to help” displayed first:
 +
** Better bounce rate, reduced by 14%
 +
** Reduced negative feedback by 74% according to Page Feedback Tool
 +
** No negative impact on number of actions or % of engaged visits
  
'''[[:en:images/a/a8/Condensed_Target_Presentation_DRAFT_(Eng)_to_post.pptx|Principal Publisher analytics]]'''
+
Questions:
*We offer A/B testing and personalization service
 
*We work with Adobe Target
 
*It’s integrated with Adobe Analytics
 
*Personalization
 
**Allows you to display customized web content to users
 
**Examples include geographic, mobile vs desktop, new vs returning users etc.
 
  
*How does A/B testing work?
+
Q: Would the H1 be more impactful if it was more than just "Raising the bar" to something like "Raising the bar to tackle climate change in Canada"?
**Traffic is randomly allocated to different variations
 
*Example - My Service Canada Account - Requesting access code
 
**Ran a test for 7 days using 4 different variations to see which had the highest click-through rate
 
  
*[https://www.gcpedia.gc.ca/wiki/Library_of_past_A/B_tests_%26_personalization_activities_in_the_GC We keep a library of past testing on GCPEDIA]
+
A: The web team came in late to this campaign, that was more with the marketing team. In future campaigns we will be involved earlier, as well as use A/B testing and adjust accordingly.  
*These tests support digital government
 
**Digital ambition
 
**Policy on Service and Digital
 
**Digital Standards
 
*This is a cost-recovered service
 
*Institutions can get access to Adobe Target to create and manage your own tests and personalizations
 
*We offer training and support
 
*Contact us: [mailto:gc.analytique-analytics.gc@servicecanda.gc.ca gc.analytique-analytics.gc@servicecanda.gc.ca]
 
'''Question''': How can we add to the library?
 
  
'''Answer''': Just reach out to us and we can add it.
+
Q: What were the top clicks on Raising the bar section and what does this tell us?
  
'''DTO - Institutional landing page'''
+
A: Most clicked:
*Launched this week. More details can be found here: [https://mailchi.mp/c83cdb09140a/february-2024-updates-for-web-managers-and-directors-fvrier-2024-mises-jour-pour-les-directeurs-et-les-gestionnaires-du-web-20598714?e=b487ea7bfa New stabilized institutional landing page template]
 
*ILP is the mandatory landing page for institutions and agencies
 
*We’ve stabilized the Beta version (available since 2021)
 
*We’ve clearly marked which elements are mandatory and optional
 
*You can reuse elements for other web purposes
 
'''Question''': Are we back to using images for Ministers?
 
  
'''Answer:''' When we took out the pictures, we saw that many departments used images anyway. Then we were specifically asked to include the images for this iteration.
+
* At home
 +
* Canada Greener Homes
  
'''Question:''' Is the WET 4 version updated with the new ILP components?
+
After this section, many users went back to “We’re here to help” section
  
'''Answer:''' Yes, everything should be good, the changes were merged into WET 4 today.
+
A/B testing strategy for Life Events on Canada.ca - Lisa Fast DTO
  
'''Discussion - Meli-melo - Principal Publisher'''
+
* Number of teams looking at life events approach on Canada.ca
*There are only a few users that have implemented the older version of meli-melo that caused problems
+
* We compare visitor behaviour analytics between version A and B
*This situation is similar to ROT - it’s up to departments to make sure content is working properly
+
* Best practices
'''Question:'''
+
** Test small changes incrementally
 +
* Test 1 Narrow Banner April 22-28
 +
** When there was a large banner, negative results
 +
** Wanted to test if narrow banner was more effective
 +
** Tested on mobile and desktop
 +
** Predicted that there would be increase clicks on links, and less use of search function
 +
** Success criteria was met, significant increase in click-through rate for most requested links
 +
** Statistically significant: When the data shows that the differences are not due to chance, when it’s due to a design change
  
It can be difficult to keep up these developments without any notice. Is there any plan to ensure that a similar problem doesn’t happen again?
+
Q: Can we test not having a banner at all?  
  
'''Answer:'''
+
A: Many pages don’t have a banner.
  
Meli melo is a way to experiment. Principal Publisher gives departments a checklist, and we’re here to help. It’s not a service that will expire that we can give notices about.
+
In this case, we had a lot of issues with users hitting spam sites. So it was important to show that the page was authentic, to reinforce trust. So that’s why the banner is used here, for branding/trust purposes.  
  
If a deprecated Meli-melo stays on the page, it doesn’t necessarily mean that pages will break.
+
In general we can say that shorter links are better, especially on mobile, because it’s easier to navigate. Especially when links wrap, it pushes the content down.  
  
'''Question:'''
+
Test 2: Shorter page with All services button
  
Could there be a heads up before moving from one version of Meli-melo to another?
+
* Testing to shorten Canada.ca homepage by replacing 12 links with an “all services” button
 +
* There were some themes that represented only 1% of users, so better to put them in “all services”
 +
* We will see how the clickthrough rate compares with and without all themes listed
  
'''Answer:'''
+
Comment:
  
The dates are indicated on the page [https://wet-boew.github.io/GCWeb/m%C3%A9li-m%C3%A9lo/m%C3%A9li-m%C3%A9lo-en.html Méli-mélo features - Canada.ca (wet-boew.github.io)]
+
I don’t really see many of these pages as services. Many pages are informational.  
  
'''Agriculture and Agri Food Canada'''
+
A: It depends how you look at it. But you’re right, perhaps “Services” doesn’t capture all the content. We might need to see if there’s a better way to describe the pages.  
*Our AgPal chatbot is back online https://agpal.ca/en/home
 
'''National Research Council'''
 
*X and new thick redline has been implemented - this is manual until the WxT Drupal file are provided.
 
'''DTO'''
 
*Reminder that the Generative AI Chatbot kickoff mtg is tomorrow. See Slack Chatbot channel if you missed it or contact [mailto:Lisa.fast@servicecanada.gc.ca lisa.fast@servicecanada.gc.ca].
 
'''Principal Publisher'''
 
*WET-BOEW office hours are every Tuesday
 
  
*Walk-in from 1pm to 2pm (Eastern time)
+
===Roundtable===
*Booking available from 2pm to 4pm (Eastern time)
+
'''TBS'''
  
*Check this link for more information: https://github.com/wet-boew/wet-boew/wiki/WET-Office-hours,-Heures-de-service-de-la-BOEW
+
* Our partners have updated their guidance for the DRRs:  
 +
** Web guidance: [[/test.canada.ca/experimental/results-resultats/drr-content-model.html|Departmental Results Report - Canada.ca]]
 +
** At a glance: [[/test.canada.ca/experimental/results-resultats/drr-at-glance.html|[Name of organization]’s [fiscal year] Departmental results report: At a glance - Canada.ca]]
 +
** Full report: [[/test.canada.ca/experimental/results-resultats/drr-full-page.html|[Department name]'s [insert fiscal year] Departmental results report - Canada.ca]]

Revision as of 15:45, 10 June 2024

Back to GC Web Priorities Meetings - Réunions des Priorités Web du GC

Meeting information

Agenda

Updates

DTO

PCO/Public Safety

  • Departments are collaborating to provide information on wildfires
  • Public Safety (PJE Theme) - Monitoring wildfire events in Western Canada. Preparing for eventual requests for assistance from provinces. Evergreen wildfires event page is being reviewed and updated by partner departments: https://www.canada.ca/en/public-safety-canada/campaigns/wildfires.html

Presentation - ECCC - Jade Bourdages - Climate action awareness advertising campaign 2023-24

  • November 2023 campaign was launched
  • Goal was to show what Canadians can do for climate change
  • Data was very valuable in making adjustments during campaign
  • Two focuses
    • What individuals can do
    • How the government can help
  • What visitors were interested in - how government can help
    • Loans
    • Incentives
    • Grants
  • 4 out of 5 top clicked links, in the “what government can do” section
  • We also used the Page Feedback Tool
    • Some people were insulted by emphasis on individual action, on such a big problem
    • 80% of feedback said that government was most responsible to solve problem
    • People want government to be part of the solution
    • Global was the 3rd most used term in feedback
  • People expect that the GC website would focus on government action, rather than individual action
  • A/B testing, we tested what was displayed first “We’re here to help” first, and the “Individual” section
  • “We’re here to help” displayed first:
    • Better bounce rate, reduced by 14%
    • Reduced negative feedback by 74% according to Page Feedback Tool
    • No negative impact on number of actions or % of engaged visits

Questions:

Q: Would the H1 be more impactful if it was more than just "Raising the bar" to something like "Raising the bar to tackle climate change in Canada"?

A: The web team came in late to this campaign, that was more with the marketing team. In future campaigns we will be involved earlier, as well as use A/B testing and adjust accordingly.

Q: What were the top clicks on Raising the bar section and what does this tell us?

A: Most clicked:

  • At home
  • Canada Greener Homes

After this section, many users went back to “We’re here to help” section

A/B testing strategy for Life Events on Canada.ca - Lisa Fast DTO

  • Number of teams looking at life events approach on Canada.ca
  • We compare visitor behaviour analytics between version A and B
  • Best practices
    • Test small changes incrementally
  • Test 1 Narrow Banner April 22-28
    • When there was a large banner, negative results
    • Wanted to test if narrow banner was more effective
    • Tested on mobile and desktop
    • Predicted that there would be increase clicks on links, and less use of search function
    • Success criteria was met, significant increase in click-through rate for most requested links
    • Statistically significant: When the data shows that the differences are not due to chance, when it’s due to a design change

Q: Can we test not having a banner at all?

A: Many pages don’t have a banner.

In this case, we had a lot of issues with users hitting spam sites. So it was important to show that the page was authentic, to reinforce trust. So that’s why the banner is used here, for branding/trust purposes.

In general we can say that shorter links are better, especially on mobile, because it’s easier to navigate. Especially when links wrap, it pushes the content down.

Test 2: Shorter page with All services button

  • Testing to shorten Canada.ca homepage by replacing 12 links with an “all services” button
  • There were some themes that represented only 1% of users, so better to put them in “all services”
  • We will see how the clickthrough rate compares with and without all themes listed

Comment:

I don’t really see many of these pages as services. Many pages are informational.

A: It depends how you look at it. But you’re right, perhaps “Services” doesn’t capture all the content. We might need to see if there’s a better way to describe the pages.

Roundtable

TBS