Collectif
Collectif.ai is an AI-driven user research tool. It automates qualitative data analysis from support tickets and interviews, offering insights and sentiment analysis to enhance customer and user experience. This helps in prioritizing product roadmaps and identifying usability issues.
Key highlights:
Overview
Collectif is an AI SaaS tool that is designed to simplify the entire continuous discovery process as much as possible by embedding results into the research flow. It is a process where research is embedded in the flow. It automatically collects basically everything of interest from support tickets, interviews, surveys, transaction results and by reading reviews.
This feedback, collected on a larger scale, enables product teams to make decisions with a much greater focus on product creation. Collectif reinforces the true value of adaptability to make it a bulletproof source of product development.
Problem
Shortly after the launch of Collectif, we put the behavior of users under observation through data funnels as well as Hotjar. Conversions dropped significantly during the onboarding phase. Many users who were new and in very high motivation to look at how the demo account worked did not do the onboarding or set up any application.
And the first assumption is that the onboarding process might be too complicated.
Research
Mixpanel events gave insight into user funnels. Later, I used Hotjar recordings to study behavior. I found gaps in usability and functionality and conducted one-on-one interviews. Through these interviews, I saw how users moved through the system to complete onboarding - how they actually did it, their challenges and pain points.
The feedback will then be processed and analyzed using Collectif.
💡 Research findings
Users do not understand the naming and meaning of the "Topics" section.
The UI is overcomplicated.
Users is distracted by various of ways to leave the onboarding.
The user does not understand the need to connect the "Analyze Reviews" feature if they need to import only "Interviews."
Users feel a lack of understanding of "what's next".
Users drop the flow because it became overwhelming to handle.
Competitive Analysis
Next, I explored the functionality and benefits of similar AI research software platforms. I compared their onboarding flows and overall products to find ways to enhance the Collectif experience. During this analysis, I noted several important differences.
Competitors offer a simpler, quicker flow. Unlike Collectif, their onboarding is linear and provides information in small pieces. They also use more general naming conventions. Completing the onboarding process is mandatory. Additionally, competitors do not require time-consuming integrations.
Deepening User Empathy
With research findings and competitive analysis in mind, I delved deeper into user decisions and needs. Even with data and an understanding of pain points, I needed to know when and how potential users discover Collectif. I decided to conduct more in-depth interviews with existing users. My goal was to understand their motivations for completing onboarding. The insights from these interviews allowed me to create a simple user persona and apply it to this specific case.
Ideation
With a clear understanding of the user in place, I moved on to brainstorming potential features for the project. My focus was on the "Must Haves"—those essential features that were both feasible and could be designed within the project's timeframe.
Structure
A framework was established through surveys, interviews, and collaborative feature prioritization within my team. It identifies three main steps for focus in a high-fidelity prototype: Personal details, Analyze selection, and Integration/Upload. This structure, combined with insights from user research, will guide future design choices.
Lo-Fi Wireframes
Next, I focused on creating low-fidelity wireframes. This step was important for visualizing the onboarding flow and making sure users have a good experience. I tried several wizard approaches to find the most intuitive and efficient process.
After extensive research and iterations, the final design came together seamlessly with a very simplistic wizard, clear step indicators, that would walk the user through with relative ease. Based on the selection in step 2, the next step is dynamic.
While two other versions offered a faster onboarding flow, they sacrificed important information about the user and disrupted the linearity of the process. The chosen design struck the right balance, ensuring a thorough and user-friendly onboarding experience without compromising on critical data collection.
Hi-Fidelity Prototype
Based on the feedback from the initial usability tests, I created high-fidelity wireframes and a functional prototype. These refined versions incorporated user insights and addressed any usability issues identified during testing.
Considerations for Long-Term Development
Taking a step back to re-examine the entire application login and onboarding cycle on a micro level will be one of my top priorities. I will pay close attention to how each concept is formulated, named, and described.
Ensuring clarity and consistency in these areas is crucial. Additionally, I will focus on using the right wording to address any hesitation users might feel during their initial tasks with the app.
Specifically, several subject areas in the current iteration will need to be worked into the user flows:
1. Provide a clearer approach to onboarding and make sure the user understands its purpose.
2. Provide gamification to engage the user.
3. Consider optional onboarding.
4. Consider implementing onboarding inside the application.
Lessons Learned
The first interaction with the application is crucial. It needs to be simple and highlight the most important functions.
Displaying the right information at the right time is essential. This keeps users engaged.
Even if a potential user is interested after exploring the demo a poor setup can quickly discourage them.
Analyzing competition and understanding the user persona can provide valuable insights. This helps refine an experience and meet user expectations effectively.