Build Once, Create More: What Building a Design System Taught Our Team

5 minute read
summary

A design system isn’t only for product teams. Tal Castillo shares the three-step process Artha used to standardize repeatable interactions across very different client projects, and why consistency, not speed, is the real payoff.

At DevLearn 2026, I shared a session called Build Once, Create More as part of the 6 Ideas in 60 Minutes session. My goal was simple: give people something they could take away and put to use right away. I wanted to share those same ideas here, along with what our team learned while building a design system at Artha.

A quick introduction if we haven’t met: I’m Tal Castillo, a proud member of Artha Learning and a 2026 Thirty Under 30 honouree. I came to learning development from the world of animation, and if there’s one thing that background cultivated in me, it’s that I love good design. Over the last while, my team has taught me something to go with it: good design at scale depends on good systems.

What is a design system, and why should L&D care?

A design system is a shared set of standards, documentation and reusable building blocks. The pieces work together so a team can produce consistent work, to the same standard, every time and at scale.

That definition comes from traditional design, so it’s fair to ask what it means for those of us in learning and development. Whether you’re an instructional designer, a developer or a one-person team, a design system does one thing very well: it standardizes the repetitive, so the mechanical tasks only have to happen once.

That matters because gruntwork is expensive. Rebuilding the same accordion for the fifth time puts pressure on the timeline, and it pulls attention away from the real purpose of the work, which is helping people learn. A good design system clears that space so we can focus on the high-value decisions. It’s about stepping back, understanding your own work deeply, and making an upfront investment of time that keeps paying you back. Build once, create more.

"But nothing we build repeats"

When I started working on our design system, I honestly didn’t know where to begin. As a custom learning company, we work with a wide range of clients and industries. That keeps the work fresh, but it also makes it hard to standardize. Content changes, modality changes, and even the most visible layer, the brand guidelines, fonts and colours, is different on every project.

So how do you standardize repetitive tasks when nothing seems to repeat?

If you’re reading this and thinking a design system could never work for your team, I understand. It looked even less likely for us. What got us unstuck was a simple process, and it’s the one I’d recommend to anyone starting out:

Assess the root problems your team is facing.

Identify the repeatable building blocks that could solve them.

Develop those blocks so they work at scale.

The rest of this post walks through each step, including what we learned along the way, so you can skip a few of our mistakes.

Step 1: Assess, and look back at everything

The first step is to pull yourself out of the weeds and look at your work at a macro level. What has caused your team the most pain? It’s usually time, scope or scale. The more useful question is why.

We looked back at everything, including the projects we’d rather not revisit. Three questions guided that review.

What took the most time? This question points you to your biggest time saver. For us, the answer was custom interactions in Storyline, D2L and Rise. Complex, multi-layered interactions and branching scenarios take real time to build.

What was repeated? Look for the elements that show up again and again. As a vendor, we could ask this at two levels: within a project (which kinds of pop-ups did this course use?) and across clients (what are we always building?). The answer to the second one, more often than I’d like to admit, was a trusty tab interaction.

What was decided? I sat down with the team and asked which decisions about look and functionality we kept having to align on. Do secondary buttons always get a hover state? Do headers always use the primary brand colour? These conversations showed us exactly what we needed to decide once and write down in our style guide.

Step 2: Identify, and get specific

By the end of your assessment, you should have a clear list of pain points. Now you can prioritize them and start solving.

Stay open-minded here, but make sure each pain point is narrow enough to act on. “Development takes too much time” is too broad. “Building interactions in Storyline is the most time-consuming” is something you can work with. That second one was a real pain point for us, with a catch: every time we rebuilt an interaction from scratch, it came out a little different. The time cost was creating a consistency problem too.

The answer can sound obvious: make building interactions take less time. How you do that is up to your team, and this is where the honest work happens. Nobody knows your process, your people and your projects better than you do.

If I can offer one piece of advice, it’s this: make it repeatable. For us, that meant templates, style guides and documentation. Before building anything, decide three things:

What is the minimum viable version?

What will make it repeatable?

Where will it live so the team can actually find it?

Step 3: Develop the building blocks

The last step sounds intimidating, but it’s crucial: build the solution.

For us, that was a Storyline library of “building block” slides. Each one is a lightweight interaction with only the components we agreed, and documented, make a good interaction. Nothing extra. Because they’re so lean, they make it possible to adapt to each client’s content and branding, while consistently keeping the same high Artha standards we’re known for.

The important part is that the decisions from the assessment phase are built in. A new project no longer starts from a blank slide and a fresh round of debates about those hover states! It starts from something the team already trusts, and the effort goes into what makes that project different.

A design system is always a work in progress

There’s one more step that never really ends: implement, then evaluate.

We’re still reviewing our system components during and after every project. We look for gaps in coverage, new pain points, and places where gruntwork has crept back in. Each project teaches us something about what the system should hold next.

So build a place and a plan for feedback into your system from the start. Knowing that feedback loop exists also takes the pressure off. You don’t need a finished system before you launch. You need a useful one, and the confidence that it will keep getting better.

Speed is the bonus

When people hear “design system,” they usually think about efficiency. That’s real, and our team feels it on every project. But it isn’t the main reason I’d encourage you to build one.

A design system is about consistency. It’s about holding your learning to a higher standard, every time, no matter who on the team is building it. Getting a faster starting point is a very welcome bonus.

Build once. Create more.

If you’re thinking about building a design system for your own team, I’d love to hear from you. I’m always happy to talk development (and great design)!

Sign up for our LinkedIn newsletter to receive updates on new eBooks, exclusive content, and the latest trends in learning and development.

Recent Blog Posts

Artha won the Training Magazine Network Choice Awards for a fourth straight year. What do teams really choose when they...

A new tool can change what is possible in learning and development. It can make a difficult production task manageable,...

Adaptive Compliance Training A consistent learning outcome does not always require an identical learning experience. Organizations can keep policies, expectations,...