articles hero

Articles in Microsoft’s Viva Engage


Viva Engage (formerly Yammer) is a platform that connects leaders, communicators, and employees to build thriving communities within organizations. Our users were interested in creating richer, more detailed content, so we introduced a type of post for users to write this more intricate long-form content. Additionally, we enhanced writing and editing capabilities for short-form posts to improve the overall user experience.



Why are we doing this?

⚡ The challenge: Creating engaging, discussion-driven content on Viva Engage is challenging. We believe long-form content will empower users to share authoritative posts that spark meaningful conversations.

We don't give users all the tools they need to create sophisticated and engaging content; we currently only enable them to create short, quick posts with limited choices. At various point in a person's career, they want to discuss different ideas, and share information in a multitude of ways. Viva Engage wants to enable people to have more say over how they create content and engage with their peers.



Corporate communicators: Corporate communicators are employees whose jobs are to disseminate corporate information to a wide audience. Some of the topics they discuss are heavy and warrant a space for discussion.


Leaders: Organizational leaders (directors, VPs, etc) want to share and discuss broad and important topics with their employees. Their aim is to disseminate ideas quickly and easily, and also receive direct feedback from those they work with.


Employees (creators): Not all employees create content, but for those who do, being able to share big ideas and format it in a way that makes it eye-catching and professional is key.



Competitive analysis

We performed a competitive analysis of similar product features. We looked at the following:

  • Facebook Workplace

  • Medium

  • LinkedIn

  • Outlook

  • Teams

  • Reddit



Insights based on analysis

Through this we identified features that my PM partner and I believed would allow users to create richer and more sophisticated content which included:

  • Accurate copy-paste actions as we uncovered that most people compose long-form content elsewhere (i.e. word, google docs)

  • Draft states

  • Extended formatting to match competitors (i.e. quotes, code blocks, etc)

  • Inline images so that media can be placed in context of the long-form content




How do we introduce long-form content?

The product manager, engineering team, and myself initially asked ourselves, exactly what is long-form content? Is it a new type of post? Is it an extension of existing types of posts? We performed some brainstorming exercises as a team to better understand our users, in hopes that it would help us make a decision that we could then test.



Exploring what constitutes long-form content

Concept 1: Only formatting improvements

We initially thought of creating additional formatting options for the existing Viva Engage publisher. Users could add in-line images, have more text options for titles, and many other things in order to create an article. However, our current reading experience paradigm did not provide a delightful experience for longer content.



Concept 2: Creating articles using SharePoint

Technologically, we thought of using a visually customized SharePoint publisher in order to create article posts in Viva Engage. This was a more conservative approach, in which the user would create a post in SharePoint and then would be able to share the post in Viva Engage. We felt that while this was technologically more simple, it would be more work for our users and a more disjointed experience.





Concept 3: Articles as a Viva Engage post type

This is the concept we ended up deciding to go with, which is creating a new article post type that is completely powered by Viva Engage. This would allow for more coherence between the new article post type and other existing post types.





Usability testing: Round 1

The product manager, researcher and myself interviewed five participants and took them through three short prototypes to get their feedback. So we went into this wanting to understand:

  1. How do people usually prefer to read articles/blog posts? Do they read them on the spot? Do they save them to read for later? In general, we wanted to understand that experience.

  2. Do people understand, based on the prototypes, how to create an article post?

  3. What are users preferences around viewing comments?

Outcome: We found that all the participants understood the creation process. We found that users often hold off on reading longer content and will open it in a new tab and read it later. They also had a strong preference for viewing the comments underneath the article as they found anything else to be too distracting.




We continued on with the design of the feature...





Until we got a new idea...

One of our new engineers thought it may be better if the creation experience also occurred in a full page, that way the user would have a better idea of how the end result would look as they're writing it. I thought this was a great idea, and so I got to work on a couple of concepts.


Concept 1: A (almost) full page modal

I created one concept in a modal experience, my reasoning was that the modal is large enough to to get an idea of what the final product will look like, but still clear that it's a modal.

User opens up article in a full page modal


Concept 2: Creation in its own page

The second concept was that creation occurs in a full page, and follows our products typical back button functionality that is consistent with the rest of the app.

User opens up article in a full page



Usability testing: Round 2

We did a second round of usability testing, this time it was an unmoderated test with 10 participants. Half the participants tested one prototype, and the other half tested another. We tested all the concepts listed above and wanted to uncover:

  1. Which model of creation best matches their mental model of how to write an article?

  2. To understand if users wanted the option to create additional post content about their article.

  3. Were there any usability issues when it came to the creation process?


Research insights
  • Both creation models performed well, but there wasn't a clear winner. While both were highly usable, we selected the full-page experience based on future capabilities and engineering feasibility.

  • Users did not seem to fully grasp the concept of writing an additional post about the article. While it is a common pattern in competitor apps, our users felt it was unnecessary.

  • Overall, participants had no issues writing, previewing, or posting an article. However, about half were unaware they needed to select a community, and it wasn’t always clear when this step should be completed. Many expected a clear decision point at the start or end of the process.





The final product






How successful was it?

Key data points (during early release)
  • Articles on average has 640k monthly active users (MAU)

  • 3,083 authors per month

  • 40.33% retention rate of authors

  • 90.05% of articles included an attachment (media or document) compared to the usual 34.42% on other post types

90.05%

Percentage of articles with attachments (media or documents)

40.33%

Retention rate of authors

26.23%

Percentage of authors writing multiple articles

Early in the feature release, articles encouraged more attachment usage due to improved capabilities and additional user control. We believe this will increase long-term engagement, especially with the use of media in posts like images, GIFs, and videos. We have 26.23% of authors writing multiple articles, which is a strong base to start with since we often find the number of creators to be minimal within each company.



Much more to do

Designing articles was a bit of a winding road in terms of process. I liked that I got to exercise my research chops; however, there are still elements I want to explore further and to get more user feedback on. Some next steps for improving the experience are:

  • Simplifying the creation process in terms of the number of steps and the order of operations.

  • Continue exploring the experience of how the article opens in a new page, that has been an area where users have conflicting and strong opinions. We have an A/B experiment planned to explore this.