(Part 1): I read a Learning Science Article and Built an Instructional Design Portfolio Project From Scratch
- Alice Ferguson
- Mar 25
- 4 min read
Here's what happened:
Part 1: Research, Storyboard & First Steps
I have a problem that many freelance learning designers probably recognise: too many NDAs, not enough portfolio.
The work exists. The thinking exists. The storyboards, the builds, the decisions - all of it is sitting behind confidentiality agreements I can't share. Which is fine, professionally. But it's a real headache when you want to show potential clients what you actually do.
So I started looking for a reason to build something from scratch - something I owned entirely, that would let me work through the full process: research, content writing, storyboarding, and then the actual build.
I just needed the right spark.
The Spark: Dr Philippa Hardman's Instructional Design AI Model Stack
I've been following Dr Philippa Hardman's work for a while now. What I appreciate most about it is that she's not talking about AI as a replacement for learning designers - she's talking about how to embed it, utilise it, and actually maximise what these tools can do.
Her investigations into the intersection of learning science and AI are genuinely interesting to me in a way that much AI discourse isn't.
So when I read her article on the AI model stack for instructional design, my first thought wasn't "I should implement this at work." It was "I wonder if I could use this to drive a whole project."
The idea: use the framework from the ground up - research model first, then creative, then reasoning, then execution - and document the process as I went. No brief from a client. No stakeholder sign-off needed. Just me, the framework, and a question I actually wanted to answer.
The Brief Came From a Snowboarding Trip (As Most Good Ideas Do)
I'd just got back from a snowboarding holiday. It was a great trip - genuinely, but my provider wasn't going to get five stars.
Nothing catastrophic happened. The mountain was brilliant. But there were a few moments where the gap between what I'd imagined and what was actually there felt wider than it needed to be. When I got back, I started wondering: could a travel provider actually close that gap?
Not by making the holiday better - that's a logistics and operations problem. But by helping guests arrive with more accurate expectations? By giving them tools to adapt when conditions change?
That felt like an instructional design problem. And I had no idea whether travel providers even need to solve it - the data on whether expectation management actually drives reviews and referrals would be theirs, not mine. But as a design brief? It was genuinely interesting. So I ran with it.
The project became Whiteout Wednesday: What To Do When The Mountain Says No - a 5-minute scenario-based microlearn for ski holiday guests, set at Les 2 Alpes in France (although I am debating on whether to generalise it?).
The Bit That Was Actually Hard
I should be honest about the friction, because it's real.
I'm a SAM person rather than a strict ADDIE follower - I like a solid content document, but I also want the flexibility to let interactions evolve as I build. The storyboard I produced for Whiteout Wednesday is detailed: 10 screens, three distinct assessment types, and consequence-driven feedback for each decision point.
But some of those interaction decisions felt harder to make before I could see the visual identity coming together. Normally, when I'm working with a client or a stakeholder, that collaboration helps. You see the fonts and colours working together, you see where the emphasis naturally falls, and suddenly it becomes obvious which interaction type suits that moment. Doing it solo meant I had to make some calls that felt slightly provisional - good on paper, but waiting to be confirmed in the build.
Some parts of the build, though, became genuinely easier. When you've documented why every decision was made, the execution feels more coherent. You're not second-guessing yourself. You're just building to a spec you actually trust.
It's an interesting trade-off. One I'm still thinking about.
What Whiteout Wednesday Actually Is
The microlearn puts the learner in a realistic storm-day scenario at Les 2 Alpes. They wake up to a weather alert. The glacier lifts are closing. What do they do?
Over five minutes, they work through three decisions: checking lift status before wasting time at a closed station, understanding why tree-lined runs work in flat light when open bowls don't, and discovering that the resort - the new pool and ice rink complex, the hidden village of Venosc accessible by gondola, the fromagerie in the valley - has a lot more to offer on a storm day than most guests ever find.
Every design decision maps to the research. The consequence-driven feedback follows Cathy Moore's Action Mapping. The advance organiser, the persistent Lumiplan shortcut, the downloadable pocket guide - all of it has a rationale I can explain and defend.
That's the thing Dr Hardman's framework gave me that I wasn't expecting: not just a faster process, but a more defensible one.
Where It's At Right Now - And Where It's Going
Whiteout Wednesday isn't finished yet. What exists right now is a complete storyboard and the first steps of the Genially mobile first micro learn build - and you can actually see it in progress here:
This is Part 1 of what I'm treating as a full case study series. The following parts will cover the build itself, the moments when the storyboard met reality and things shifted, and whether the finished piece does what the research said it should.
If you're a learning designer who also lives in the land of NDAs and thin portfolios, I'll be documenting the whole thing as I go. Follow along - it's a work in progress, and honestly, that's the most interesting bit.
Alice Helen Ferguson is a freelance Learning Designer based in Falmouth, Cornwall. She specialises in learning architecture - the point where good design and good learning meet - and works with agencies and L&D teams to create training that people don't resent.
Comments