Skill Spotlight
Role
UX/UI Designer
Product
Added Feature on LinkedIn
Duration
May to July 2023
Tools
Figma
Overview
I introduced an innovative feature called Skill Spotlight, which allows users to go beyond words and visually demonstrate their abilities using links, images, videos, and posts. This feature simplifies recruiters' searches by enabling them to find candidates with the most relevant assets. With Skill Spotlight, I made the job search process more engaging and effective for both users and recruiters.
I was the sole designer working on this project. It was an project directed at improving the capabilities of a website. I saw potential in LinkedIn as a career social media site.
THE PROBLEM
LinkedIn users are facing challenges with the effectiveness of the skills section, leading them to rely solely on the experience section. This poses a significant issue for career changers and individuals new to their roles who may possess valuable transferable skills despite limited experience. The underutilization of the skills section represents a missed opportunity for professionals and a waste of valuable resources for the company.
The Process
Skill Spotlight can stand apart from the crowd by facilitating the exchange between recruiters and professionals.
I conducted a competitive analysis of four platforms—Facebook, Twitter, Reddit, and Indeed—since my preliminary research revealed their diverse roles in job posting and search. Facebook features job groups and posts; Twitter offers lists, TweetDeck, and search tools; Reddit hosts career-related discussions; and Indeed showcases job listings and articles. This analysis aimed to identify gaps in the current job market offerings, focusing on enhancing networking, information exchange, and collaboration among professionals, beyond existing platforms, through the advanced skills feature.
Competitive Analysis
Key Takeaways
The LinkedIn (LI) advanced search feature could stand apart by the fact that both recruiters and professionals can make use of the skills listings
A private mode for skills search and postings could be helpful (for posting, only the other party that you are offering your skillset to could see it)
For those that want to showcase their skills and past work there could be badges/ recommendations.
The advanced skills feature needs to be not just about applying to jobs, but also connecting with other professionals and recruiters in your network
Summary
Building connections and gaining information is an important aspect of all of these social communities, however the LI advanced skills feature can take it a step further by facilitating an exchange of not only information, but also useful projects with other professionals and key hiring characteristics with recruiters. Allowing an opt in and private mode could also provide for more control.
Recruitment
Broadening rather than limiting study participation is always ideal.
.png)
In the recruitment phase, I sought LinkedIn users with skill section familiarity, aiming for 1-2 recruiters and experienced professionals. Despite speaking with 7 participants, including 1 recruiter and 6 professionals, the unavailability of the recruiter view hindered full goal attainment. Nonetheless, insights were gathered on skill feature usage, platform preferences, pain points, and hiring assumptions. Future recruitment could benefit from a broader participant pool beyond strict criteria, potentially encompassing diverse perspectives.
USER INTERVIEWS
Users lacked an understanding of the purpose of the skills feature and spoke about the enhancements they would prefer.
I conducted interviews with 7 participants within the target audience to explore current usage of the skills feature and gather insights for potential improvements. The research aimed to uncover how professionals and recruiters utilized the feature, assess user understanding and value, and gather suggestions for enhancement. Despite encountering some participants with limited knowledge of the feature, the interviews revealed misconceptions and misuse. Although challenges like social desirability bias affected responses, valuable insights were gleaned, such as the significance of skills in search results and the desire for showcasing expertise through assets or visualizations. Future improvements involve refined screening, non-leading questioning, and an emphasis on understanding participants' baseline knowledge.
affinity map
Asking directed questions while creating an affinity map creates clarity.
Upon formulating two key questions - "How is the feature currently understood?" and "What improvements do people want to see?" - the process became more manageable, resulting in the formation of distinct clusters. The analysis revealed that the skills feature served purposes such as endorsement, filling experience gaps, showcasing learning, and influencing hiring decisions. Participants also expressed desires for A.I.-driven skill listings, career track grouping, and profile enhancements. Collaboration in affinity mapping proved valuable, suggesting potential benefits in future teamwork for diverse insights and approaches.
IDEATION
Creating POV and HMW statements for dual user groups proved to be challenging.
I developed "point of view" and "how might we" statements to refine my problem-solving approach, facing challenges due to addressing two user groups. I formulated initial statements aimed at improving recruiters' use of the skills feature and assisting emerging professionals in showcasing skills for desired careers. Although not perfect, they served as a foundation. This led me to various "how might we" statements, eventually landing on "How might we help users demonstrate expertise in specific skill areas?". The process highlighted the difficulty of avoiding premature solutioning, particularly with preconceived hypotheses. To maintain open possibilities, I intend to adopt a more flexible focus in future endeavors.
Skill Spotlight can help recruiters and professionals demonstrate their skills visually and streamline the recruitment process.
The solution development process involved numerous iterations and brainstorming. Starting with an initial "how might we" statement focused on showcasing skills, I explored concepts like skills visualizations and a profile spotlight. Refining the statement to enhancing recruitment for recruiters led to AI-powered recommendation engines and rubric matching. Eventually, I settled on a new "how might we" statement centered on proving experience in specific skills, leading to the idea of skill assets for representation. This process emphasized the significance of clear problem definitions and "how might we" statements, aligning with the notion that a well-framed challenge is halfway solved, as emphasized by the Interaction Design Foundation. In future endeavors, I plan to employ helpful templates, such as the user, need, and insight framework for crafting point of view statements.
personas
I created dynamic personas to add depth and focus to the feature.
I created two personas- one for the recruiter and one for the professional. I chose this method so that I could really bring the data to life. It gave my design a consistent focus. The result was that I came up with Rob the Recruiter and Patricia the Professional. Rob mainly wants to get an idea of a candidates’ skills in one snapshot, while Patricia wants to show off her knowledge despite her lack of extensive experience. I learned that personas can really help guide your work, but they also have to be well designed for them to be useful. I think I originally had very vague goals, frustrations, and scenarios for each persona. Next time I will consider the scope I want my personas to have before creating them.
task and user flows
The task flows and user flows helped me clarify functionality and the direction for my designs.
I created task flows and user flows for my project in order to map out how users could benefit and the steps they would take to accomplish tasks. As a result, I was able to get a clearer picture of the functions that the website needed to have and the processes needed to look like before I started designing. I learned about the need for writing out your use cases before creating flows. Doing this instead of jumping straight into the diagrams organized my thoughts.
Low fidelity wireframes
The low-fidelity wireframes helped me to iterate on, but stay true to, LinkedIn's designs.
I created low fidelity, paper wireframes for the website. I did this so that I could make quick and easily revisable drafts of what the website would look like. This time because I was working with an already existing website I placed my sketches next to the screenshots to be sure I would stay true to the original design. I was able to get a visual representation of the user flows I had created previously. I achieved my goal because I could tell where to concentrate my efforts when making my high-fidelity prototype. I learned that I should have started by looking at Linkedin’s design patterns first because it would be necessary to stick to them as closely as possible. There were a lot of elements I still had to change at the hi-fi part because of this oversight.
high fidelity wireframes
Using an existing design system can be efficient, yet limiting.
I started creating the high-fidelity prototype. I chose to go to this point instead of having a mid-fidelity wireframe because I was working on a product that was already laid out. Therefore, there wasn’t as much need to slowly work up to each granular detail. Besides the need for a component library, there was not as much need to create elements and add images from scratch. The result was a clickable, functional prototype that users could go through for testing. I learned about both the benefits and limitations of working within an already existing design system. On one hand it can simplify and streamline the process because you don’t have to design every element, but on other hand it can be somewhat limiting because things you would like to improve cannot always be changed. It is also hard when not all design elements are published online. Companies like IBM make their systems available to the public, but LinkedIn doesn’t.
USER TESTING
The results from the user testing determined the priority of revisions and led to an improved design.
I conducted user testing and acted upon the insights from testing to make priority revisions. I did moderated user testing because I wanted to see participants’ reactions as they went through the prototype. I also wanted more commentary and to be able to ask follow-up questions. I used AirTable to aggregate all the data from the user testing and analyze which features needed to be changed most. Overall task completion rate was 100%. All users were able to accomplish the goals set out for them in the user testing. The average number of errors during each task was as follows: 1 for the adding an asset flow, 0 for the adding skill tags flow, and 2 for the skill search flow. The average time to complete each task was around 1 minute for adding an asset and skill tags, however it was about 2 minutes for skill search. The average task rating on a scale of 1-5 (5 being easiest and 1 hardest) was 4.86 for both adding an asset and adding skill tags, however it was 3.67 for skill search. The main problems were in the skill search flow. Users were confused about why skills were not listed under people and why the page didn’t respond to their search query of “ui design”. This could be due to a multitude of factors such as the fact that users were asked to act as recruiters during this flow, which may have been unfamiliar to some of them, or it could be due to actual design flaws. To address the second possibility, I made revisions.
I ended up leaving the skills search as its own separate entity and not classifying it under people search because within LinkedIn’s current framework that search is used for looking up the names of individuals and not the skills that they possess. Therefore, skills would need to have its own search as it is a new feature. I was able to get rid of some of the confusion by removing the first screen and creating it as if the user had already entered their search, rather than having to click the search bar. I made several other UI based changes that can be seen below. I learned a lot about the need clarity when it comes to creating research guides, asking users questions, and designing overall. All three are key to making it obvious which steps the user needs to take and getting proper feedback to implement in your designs.
The Outcome
The results stated above would be some of the ones I would look at if I were to evaluate the efficacy of this design. Other measures to look at could be adoption rate, retention rate, and customer satisfaction. Adoption rate and retention rate are particularly great to look at after adding a new feature because you can see whether new users are being obtained as a result of your design and whether those users are staying long-term after a change has been made. If you see a drop off in either of these measures, then that would be a sign that a change needs to be made or that the feature is not working as it should. The customer satisfaction score is also a great measure of whether users are satisfied with a product. This is especially important after you make major changes to a product.
The Solution
Asset Upload - Ability to upload proof of what you’ve done with a title, summary, and image, video, etc.
Adding a Skill Tag - Ability to tag your current posts or create new posts will skill tags that identify what the primary skill used in that post was. For example, if you earned a certificate you could tag it with the skill you learned.
Skill Search Function - Allows recruiters see candidates listed out by their number of assets, view their profiles, and see evidence of their skills.
Skill Spotlight made the skills section useful. If this feature were to be implemented it would mean that skills are no longer just a laundry list, but a highlight one’s abilities and make them more visible to recruiters that can get them hired. This would not just be exclusive to those with work experience, but people with any kind of demonstrable skill could take part.
Lessons
I learned that the major social sites are not really focused on skills listing as much as they are on using connections to find roles. As for the specifically career focused sites like LinkedIn and Indeed they do have skills listings, but the skills don’t really serve much of a purpose besides being highlighted when applying for jobs. At that point if you don’t list a certain skill the site will ask if you’re sure you want to apply without having that skill listed or on the premium version of LinkedIn you can actually see which particular skills you are missing. Either way, none of the available sites show proof that you possess these skills. There is a very specific target user for this type of feature. Career changers, entry level professionals, new graduates, and recruiters could benefit. These are people who either would not have extensive, relevant experience to list or would want another way of verifying candidates experience.
I also learned that a finished project is better than a perfect project. The recruitment and ideation process slowed down the project significantly. Recruitment was difficult because I was focused on a specific subset of LinkedIn users. I should have focused on anyone using the platform, even if they were not a recruiter. I had several ideas, but I was not always thinking of the right question to answer. My best HMW was actually quite straightforward and I think it helped me come up with the best solution.
