All work

        Chegg·Skills gap & career readiness·Education
      

Turning a job board into a learning-to-employment system

Chegg Career Center could show a student thousands of openings and nothing about how to become the person who gets one. I led UX for its evolution into a skills-driven readiness platform, where a career names the skills it needs, each skill resolves to something you can learn, and the chain ends at a real internship or job.

case.md

$ cat outcome

engagement: +20 to 35%

early_drop_off: -18 to 30%

Chegg Career Center career spotlight for Interior Designer showing average salary, education level, a selectable list of required skills, online courses teaching the selected skill, and related internships and entry-level jobs — The whole argument in one screen. A career spotlight names its required skills, selecting a skill loads courses that teach it, and below that sit the internships and entry-level jobs the skill unlocks. Browse became a path.

Why it mattered

Students could see the jobs they wanted but not a path to qualifying for them.

60-second read

Problem

Career Center was a fragmented job board. Students could browse openings but nothing connected what a job required to what they were missing or what they could do about it.

System decision

Make the skill the unit that joins everything: a career names its required skills, each skill resolves to courses, and the chain lands on internships and entry-level jobs.

My contribution

Led UX end to end across web and mobile with a team of six. Information architecture, the skills-gap model, personalization, and the component system.

Impact

Career preparation engagement up 20 to 35%, profile and résumé completion up 15 to 25%, early-session drop-off down 18 to 30%.

01 · Shipped

A connected system across web and mobile: over 600 career profiles, a course catalog, an internship marketplace, and a personalization layer that gave them a reason to talk to each other.

20 to 35%

increase in engagement across career preparation flows

15 to 25%

increase in profile and résumé completion

18 to 30%

reduction in early-session drop-off

02 · The problem

Chegg reached students at the moment they were deciding what to do with their lives, and answered with a search box. Careers lived in one place, courses in another, internships in a third through a partner. A student could browse all three and still not know what to do on Tuesday.

The gap was not content. Nothing in the product connected a job's requirements to a student's actual position, so every screen was a dead end dressed as a destination.

The original Chegg Student Hub homepage with a search field and three cards for cheap textbooks, homework help, and internships and jobs — Before: careers as one of three unrelated errands, sitting beside textbook rentals.

The original Chegg internships dashboard with keyword and location search fields and a list of internship suggestions — Before: the internship surface. A keyword box, a location box, and no notion of what the student was qualified for.

03 · The unit that
joins it

I made the skill the joining unit. Once a skill is a first-class object rather than a keyword, every relationship falls out of it: a career requires skills, a course teaches one, an internship tests them, and the distance between a student and a job becomes a list of specific, closable gaps.

              CAREER
              
              SKILLS REQUIRED
              
              COURSES FOR THAT SKILL
              
              INTERNSHIPS
              
              ENTRY-LEVEL JOBS

Chegg Explore Careers browse page with faceted filters for major, strengths, interests and salary, a grid of career cards, and a career wish list carousel — Discovery, faceted by what a student can actually answer about themselves: their major, what they are good at, what they like, and what they need to earn. The wish list turned browsing into a set of saved intentions.

Chegg Browse Courses pages with category, skill, level and price filters, a course detail page, and a course in progress showing its lecture list and completion percentage — The course side of the loop, filtered by skill rather than subject. A course detail page names the skills it teaches and the careers those skills lead to, closing the circle back to the spotlight.

Chegg internships flow: an entry screen, school and major questions, matched internship suggestions, a search dashboard, and an internship detail page — The internship path, powered by the Internships.com partnership. Two questions about school and major before anything else, because a match is worth more than a search.

04 · Personalization

Rather than infer intent from behavior a new student had not produced yet, I asked one question and branched on the answer. Three postures, each with a different first screen: exploring options, focused on an internship search, or working on skills.

Internally we called them rookie, pro, and superstar. The label mattered less than the fact that a first-year student and a graduating senior no longer got the same homepage.

Wireframe map of the personalization flow branching from three student intents into a personalized home, with search, menu, and category, browse and detail templates for internships, courses, careers and advice — The architecture, mapped before any pixels. The teal block is the personalization branch, the yellow is the personalized home and the four content spines it feeds. Every spine repeats the same category, browse, detail rhythm, so the product only has to be learned once.

Chegg Career Center personalized dashboard on web and mobile showing résumé completion at 75 percent, application counts, enrolled courses, wishlist items, and top recommendations — The dashboard a returning student sees. Résumé completion as a progress ring, applications and enrolled courses as counts, and recommendations tagged with the interests that produced them plus an Edit control, so the personalization stays legible and correctable.

05 · Mobile

The app was not a reflow of the site. It assumed short sessions between classes, so it front-loaded the value proposition and reduced the core interaction to a judgment a student could make in a second.

Chegg Career Center mobile onboarding: a splash screen followed by four panels reading your path to a job starts here, explore careers, build your skills, and land internships, each with sign in and join now — Onboarding as an argument, not a tour. Four panels that each name a concrete promise, with Skip available on every one.

Chegg Career Center mobile screens showing interest selection chips, a swipe left to remove and swipe right to save instruction card, recommendation cards, and a current status selector — Interests as chips, then swipe to save or dismiss. Each swipe is training data, so the feed improves through use rather than through a settings screen. The status question resurfaces in context when a student's situation changes.

Chegg mobile course screens showing a course overview with lectures, progress and enrollment, and a course in progress — Course consumption on the phone, including progress state, because the skill-building half of the loop had to work in the same short sessions as the browsing half.

06 · Making it
reproducible

A system this size, built by six designers, holds together only if the parts are specified rather than improvised. We defined the components and the brand voice up front, so consistency was the default rather than a review note.

Chegg button specification sheet showing large, small and mini buttons in primary, dark gray, light gray, flat, outline and orange variants, each with default, hover and disabled states plus exact typeface, size and hex values — Buttons, fully specified: three sizes, six variants, three states each, with typeface, pixel size, and hex value written on the sheet. Unglamorous, and the reason six designers could ship a coherent product.

A brand attribute workshop, run with printouts on a wall. Moxie, serendipity, benevolent, no BS, savvy. Sorting real screens under each word is how you find out which ones the product is not living up to.

Chegg visual direction board with headline and header type specimens in Avenir Next Condensed, primary and secondary button styles, flat icons, a color palette and student photography — The visual system that came out of it: type scale, button styles, palette, iconography, and a photographic direction that looks like actual students rather than stock models.

07 · First time
leading

Chegg was the first team I managed, six designers across web and mobile. I learned the obvious lesson the hard way: I could be the person who solved the interesting problems, or the person who raised what the team could attempt, but not both.

What worked was defining the system, the skills model, the component specs, the shared vocabulary, then handing over real problems inside it rather than safe ones. It is the same instinct I have used in every role since, including as a principal individual contributor.

Wireframe map of the account area showing menu, my account and about screens, plus my internships résumé, applications and wishlist, my courses enrolled and wishlist, and careers wishlist — The account area, specified down to the tab states. This is the kind of surface that gets improvised when a system is not written down, and the kind six people cannot build consistently without it.

08 · Measured

Career preparation engagement up 20 to 35%

Profile and résumé completion up 15 to 25%

Early-session drop-off down 18 to 30%

Cross-surface retention, web and mobile increased

Drop-off is the number that tested the actual bet. Asking one honest question instead of inferring intent was supposed to stop students bouncing in the first few seconds, and it did. Résumé completion is the one I would defend hardest, because a student only fills in a profile when they believe what is on the other side is worth the effort.

09 · Takeaways

01 A job board lists opportunities. A readiness system closes the distance between where someone is and what they want, which is a different product.

02 Pick the unit that joins your domain. Here it was the skill, and once that was settled the information architecture mostly wrote itself.

03 Personalization should ask one honest question rather than infer from behavior the user has not produced yet.

04 Design the returning state, not just the empty one. A student with nothing entered still needs a reason to come back.

05 Leading a team for the first time taught me to hand over the interesting problems, not just the safe ones.

A list of openings is not opportunity. The path to one is.

The Admissions vertical