Fidelity Learning Center
A research-tested learning model for new investors
Client
Fidelity Investments
Role
Lead designer, UX and visual design
Team
Design director, content design and strategy, UX research, product, engineering
Timeline
2 years · 2015–2017
Overview
The Fidelity Learning Center is where customers and prospects go to learn about investing, through articles, videos, webinars and courses. In 2015 Fidelity asked our UX design team to get more people using it and coming back. I was the lead designer for UX and visual design, working with a design director, content strategists, UX researchers, product owners and developers. Over two years we designed short, guided lessons to replace long courses, starting with the hardest topic we had: ETFs. The design went into development in 2017 but hadn’t launched when I left, so every result here comes from research, not a live site.
470
customers and prospects in testing
94%
noticed the check-in questions
6 → 1
concepts narrowed to one model
Source: Fidelity UX Research online studies (UserZoom), October 2015 (207 people) and February 2016 (263). Research results, not live-site metrics.
01
Problem
Fidelity’s Learning Center had plenty of content, but most visitors read one article and left, because long, jargon-heavy courses felt like homework.
The brief was to increase return visits, make learning more guided, and make complicated topics approachable with shorter content.
One article and gone. Repeat visits were low. Most people left after the first article and rarely reached a second.
Courses felt like homework. They were long and hard to finish, and in earlier research customers saw them as little more than a stack of articles.
Built around products, not people. Articles were long and full of jargon, content was sorted by product, and a “learn more” link from a tool sometimes landed on a page that didn’t match the topic or level.
02
How six concepts became one learning model
We worked lean: write the brief, list our assumptions, sketch, and test with customers along the way. Five decisions shaped the model:
Start with the hardest topic. We chose ETFs as the test case. If a short lesson could make ETFs clear, it could handle most of the Learning Center.
Go wide, then test. I designed six concepts, from Khan Academy-style courseware to a personalized wizard and card-based bursts of information. Internal reviews cut them to three: Bootcamp (short and coach-like), Courseware (a structured curriculum) and Module (an editorial mix of formats). In October 2015, 207 customers and prospects tried all three.
Combine the two winners. Module was rated most engaging and understandable; Bootcamp was the least cluttered. Neither was enough alone, so the final model pairs Bootcamp’s focus with Module’s variety.
Keep the conversation two-way. Each short section ends with a quick question. A wrong answer becomes a short explanation instead of a dead end, and check-ins let people keep going, take a break or switch topics. A progress bar shows how much is left.
Drop the school talk. Anything that sounded like a course, a curriculum or a graduation tested as too much work. We tested names for each block of content and used the one people preferred: “section.”
Where we changed course
I started with courseware because it looked like serious learning, and customers liked it least: it came last in preference and had the lowest task accuracy. Articles were the format people wanted most. The first ETF concept was a 14-day challenge, five minutes a day; we kept cutting until it became a single five-part challenge you can finish in one sitting.
The ETF Learning Challenge (late 2016) on desktop and mobile. Each short section ends with a check-in question, and right or wrong, the answer comes with a one-line explanation so people keep going.
Inside a section. A sticky bar tracks progress through the five parts, so a short lesson always shows how much is left.
03
Outcome
Validated in a second round. In February 2016, 263 customers and prospects tested three versions of the challenge. 94% noticed the check-in questions and 78% the progress bar, and both rated as helpful for learning.
Designed and handed off. I delivered the ETF Learning Challenge with responsive layouts, a style guide and redlines in December 2016, and it went into development in 2017. That year we also extended it into learning paths for Fidelity’s ETF center. It hadn’t launched when I left in January 2018.
A playbook for every topic. With our content designer, I wrote the Active-Engagement Learning Model Playbook (January 2017): how long a topic should be, how to title sections, when to ask a question, and a template any Learning Center topic can follow.
04
What I learned
People want answers, not courses. Most visitors arrive with one quick question. Small sections that add up beat a curriculum that asks for an hour.
Let the test pick the hybrid. No single concept won. The best version took the strongest part of two.
Write the rules down. A playbook let the model outlive one ETF lesson and gave content teams a way to build new topics on their own.
