Learning Design

Learning design ยท Robotel ยท UBC ยท Community

Learning that fits the learner, from analysis to SCORM

I design and build e-learning end to end: needs analysis, objectives, storyboards, scripts, Rise and Storyline builds, accessibility checks, SCORM and evaluation. It started in an art room, with children who each learned in their own way.

Roles
Learning experience designer and developer
Tools
Rise 360, Storyline, H5P, HTML/CSS/JS, SCORM
Frameworks
Bloom's, cognitive load, UDL, 70:20:10
Education
M.Ed., Educational Psychology / Special Education, UBC
This folder has two parts

Start with the case study. If you'd like to see how I came to learning design, and what else I've done in it, read My path.

01 ยท The brief

An internal course that changes how people work

At Robotel I designed and built an internal course for staff. It was my project from the first interview to the SCORM package.

About this example. I built the real course with real learners, real data and real reviews. Because it is confidential, this page rebuilds it on a different topic, Accessible by Default, using the same process, tools and decisions. The people, quotes and numbers are sample data, kept very close to what I worked with on the real course.
Request

"Staff who build activities keep publishing content some students can't use. We need everyone to get it right without slowing down."

Audience

About 20 staff who create or support SmartClass activities: content creators, customer success and developers.

Constraints
  • Self-paced, under 40 minutes
  • Works on desktop and phone
  • Tracked in the company LMS
  • Six weeks, one subject-matter expert
My role
  • Analysis and learning objectives
  • Storyboards and scripts
  • Build in Rise 360 and Storyline
  • Custom HTML/CSS, SCORM, evaluation

The process

I followed ADDIE, with a review at every stage instead of one big review at the end. Each stage produced something the manager and the subject-matter expert signed off.

02 ยท Analysis

First, is training even the answer?

Before writing a single screen, I looked at who the learners are, what they do now, and why. Some gaps needed a course; others needed a better template.

Four ways I gathered data

Interviews

  • Manager: the business goal
  • Accessibility lead (SME): the rules
  • Five staff: how they build today

Survey

  • 15 staff rated their confidence
  • Five skills, scale of 1 to 5
  • Two open questions

Content audit

  • 25 published activities
  • Checked against an 8-point list
  • Counted each type of issue

Task analysis

  • Watched three staff build an activity
  • Wrote down every step
  • Marked where errors start

Task analysis: where errors start

Building one activity takes six steps. Accessibility problems enter at three of them, so the course teaches those three moments.

Who the learners are

Three learner profiles came out of the interviews. Select one to see their starting point and what the course must do for them.

What the data showed

The survey and the audit pointed at the same weak spots. Switch between them.

Gap analysis: training or something else?

For each gap I asked what causes it. Training fixes missing knowledge or skill. It doesn't fix a missing tool or a template that makes the wrong thing easy. Select a cause to filter.

Result: three of the six gaps went into the course. The other three became a template change, a contrast tool in the editor, and a peer-review step, which I handed to the product team.

03 ยท Objectives

Objectives I could actually test

One performance goal, five enabling objectives. Each uses a verb you can observe, so each can be assessed.

Performance goal

Staff publish SmartClass activities that pass Robotel's 8-point accessibility check on the first review.

Mapped to Bloom's taxonomy

Select a level to see which objectives sit there. Most sit at Apply, because the job is doing, not remembering.

Alignment: objective, practice, assessment

Every objective has a practice activity in the course and a quiz item that tests the same verb. Select a row to follow one objective through the course.

Why it matters: because every quiz item maps to one objective, the quiz report later showed exactly which skill needed more work.

04 ยท Design strategy

Theory, turned into decisions

I don't name frameworks for decoration. Each one changed something concrete in the course.

70:20:10: the course is only the 10

Most learning happens on the job. The course starts the change; the job aid and peer review make it stick.

70 ยท On the job

A one-page checklist used on every new activity for 30 days.

20 ยท With others

Peer review in pairs, plus a chat channel for questions.

10 ยท Course

Five short lessons and a final check.

Course map

Five lessons of 4 to 8 minutes, one skill each, plus a five-minute final check: under 40 minutes in total. Experienced staff can test out of lesson 1.

05 ยท Storyboard & script

Everything on paper before anything in Rise

A storyboard is cheap to change. I got sign-off on every screen and every line of narration before building.

Storyboard

Each row describes one screen: what learners see, hear and do, plus notes for accessibility and the build. Pick a screen from lesson 2.

Script and review history

Narration went through three versions. Move between them to see what the SME and the pilot changed, and why.

RemovedAdded

Review rounds

Result: five review rounds, each with a sign-off, so the build started from an approved script and almost nothing was rebuilt.

06 ยท Build

Built in Rise, extended with code

Rise handled the structure. Where it fell short, I added a Storyline interaction and my own HTML, CSS and JavaScript.

Click through the course below: the lessons, the blocks and the interactions all work. Switch the theme to see the brand styles I added to the exported package.

Which tool for which job

Rise 360

  • Lessons, text, process and accordion blocks
  • Knowledge checks and the final quiz
  • Responsive on phones with no extra work

Storyline

  • The branching scenario in lesson 5
  • Variables to track choices
  • Embedded as a Storyline block

HTML, CSS, JS

  • The live contrast checker in lesson 4
  • Brand fonts, colours and focus styles
  • Added to the exported SCORM package

A look at the code

Two snippets: brand overrides added to the exported package, and the heart of the contrast checker.


  

Result: one course, three tools and a branded look, all inside a single SCORM package the LMS could track.

07 ยท Assessment

Questions about the job, not about definitions

Apply-level objectives need apply-level questions. Most items show a real activity and ask what to fix.

Final check

  • 10 items drawn from a bank of 20
  • Pass mark 8 of 10, unlimited attempts
  • Each item tagged to one objective

Feedback

  • Explains why, not only right or wrong
  • Points to the lesson to review
  • No time limit

Item types

  • Scenario multiple choice
  • Choose the best alt text
  • Spot the problem in a screenshot

Try four items

08 ยท Accessibility

A course about accessibility has to be accessible

I tested the course the way learners would meet it: keyboard only, with a screen reader, on a phone and zoomed to twice the size.

Tick off the checks. Each one says how I tested it and what I changed.

Result: the course passed all eight checks before the pilot, and the pilot found only one accessibility issue, the drag-and-drop, which I replaced.

09 ยท Pilot & launch

Five people tried it before twenty did

A pilot group took the course in Review 360. Their comments became an issue log, and each fix was checked before launch.

From Rise to the LMS

I exported a SCORM package, tested it in SCORM Cloud first, then uploaded it to the LMS and checked that completions reached the reports.

10 ยท Evaluation

Did it change the work?

I planned the evaluation with the objectives, using the four Kirkpatrick levels. Completion alone proves nothing, so I looked at the activities people published afterwards.

11 ยท Reflection

What this project taught me

The best fix is sometimes not a course

Half the gaps came from templates and tools, not from what people knew. Saying so early saved weeks of building the wrong thing.

Storyboards save the build

Every change made in the storyboard was a change I didn't have to make in Rise, and reviewers could comment before anything was built.

Code widens what a tool can do

A few lines of HTML, CSS and JavaScript gave the course a live practice tool and the company's look, beyond Rise's built-in blocks.

Measure behaviour, not clicks

Completion rates said little. Auditing the activities people built after the course showed what had actually changed.