"I have an idea for a new feature or improvement to a product I use"
The concept of Ideas was to build discussion forums where customers could post their feedback on a product, then be heard and updated by the product team. Internally, this would effectively allow teams to crowd-source improvements directly from their own customers. There were two halves to this vision: the C1 (maker) experience, and the C2 (front-end, consumer) experience. I owned the majority of the C2 experience.
Ideas was slated to replace tools being used internally at Microsoft, such as UserVoice. After this, we planned to market it to other companies to use with their own customers.
Early research: what we learned
Before anything went into code, I worked together with a fellow designer to conduct interviews with PMs at Microsoft to understand and identify the issues with the current products they used. These sessions produced valuable insights on where our competitors were struggling:
• Existing tools require too much effort to successfully filter out clutter and low-quality posts
• Managers find it hard to search and sort through content (this creates a negative feedback loop with the previous point)
• This feedback loop damages the experience for customers who feel they are being ignored or drowned out by low-quality content
In some cases, there were tens of thousands of ideas that were never read or responded to.
Screenshot from a Microsoft forum on UserVoice
MVP: focus on organization
One major issue in competing products was the hierarchy of forums. Categories were separated into siloed forums. If a customer posted their idea under the wrong category (which was common due to poor naming conventions), it would never be seen by the right stakeholders and would thus never be acknowledged.
I began by rethinking the entire structure. Categories would no longer exist as siloed forums, but simply filters under a higher-level topic. Misplaced ideas would not be lost.
Note: the design below was taken from a Q&A forum, which uses the exact same formatting as Ideas.
Benchmarks: measuring the experience
Once the MVP designs were built, our UX researcher conducted benchmark tests to establish the baseline usability of the experience. I was proud to see our initial generative research and focus on simplicity pay massive dividends – our results were overwhelmingly positive and far above standard for our division. Most importantly, we still found several areas to improve based on the feedback.
+ “I’m fairly impressed with it. Having used different forums and stuff before...I was impressed how simple it is and intuitive it is."
+ “Just everything is..it just seems much more faster, professional, effective, and easy to use, compare to what I’m using."
– “I think it might be a little bit small [at the top]...it will be better if it moved a little bit lower.”
– “It looked dated. It just look old to me – the layout, the font, it looks like an old forum design to me.”
Post-benchmark: improving search
Based on research findings and our current roadmap, our team prioritized several improvements to search. The key problems that I needed to be address were:
1. Improve visibility of the search bar
2. Encourage customers to vote on existing ideas before posting a new one
3. Provide better moderation tools for PMs
1. Improve visibility of search
To improve visibility and quickly let a customer check if their idea already exists, search was emphasized up-front and center on the home page.
2. Suggest related ideas when customers post new ones
To further incentivize searching and voting for existing ideas rather than posting new ones, I designed a way to suggest related ideas when a customer begins typing a post. I believed this would help achieve our goal of reducing clutter while driving engagement with existing posts.
3. Provide better moderation tools for PMs
PMs would inevitably need to be aware of and be able to take action on duplicates, spam, and low-quality content. To accomplish this, we allowed customers to flag posts, and I designed interfaces allowing PMs to manage posts.
Communities & branding
Throughout this entire process, I needed to be mindful of our vision on how Ideas would eventually expand to a broader Communities product, including Q&A. As such, I was asked to create a proof-of-concept on how Ideas and Questions could work together.
In short, I believed Ideas and Questions should exist as pillars under a single Communities site. Each type of forum would use the same basic format, with only minor changes to content and design. I designed a new Communities home page to encompass everything.
Furthermore, we needed to be mindful of how other companies would integrate this product into their own websites. I designated a range of customizable elements that would be controlled by the C1 product, and I iterated on the navigation to make sure it would not interfere with that of third party websites.
Browsing the community
Various examples of branding
Fully responsive & accessible
The Communities site was designed with every device in mind, and with a strong focus on accessibility.
In restrospect
I was reassigned to another team before I could see this product through to public preview, so I wasn't able to gauge the full outcome. Looking back, this project was an incredible learning experience for me. Starting with a strong foundation in UX research really helped me hone in on the key problems at play and deliver a simple, intuitive, and useful tool for customer engagement. I also learned a great deal about accessibility standards that will reshape my approach in the future.
I wish I could have seen Communities reach maturity, but I'm proud to say I got it off the ground.
