We make complex things feel human.

When I was tasked with building a training module for frontline workers supporting women from culturally diverse backgrounds, I faced a real challenge: how do you create human connection in a digital-only format?

The problem with traditional eLearning

Most eLearning for the social sector looks the same — slides with bullet points, stock photos, a voiceover reading the text. It's not engaging. For communities where trust is built through human connection, it's particularly ineffective.

Why I chose AI avatars

AI avatars allowed me to create a consistent, human-looking presenter without the cost of full video production. I could produce multilingual versions with the same visual presenter — making training feel familiar across different language groups.

What I learned

• Avatars work best when the script sounds like a real person talking — not a document being read aloud
• Learners responded more positively when the avatar mirrored cultural cues in tone and pacing
• The technology freed up budget for better instructional design and real interactivity
• Always test with a sample of your actual audience before full production

One of the most common mistakes organisations make is treating online training as a direct translation of in-person training. The result is almost always worse than both formats at their best.

The fundamental difference

In-person training is fundamentally social. Online training offers something in-person never can: infinite repeatability, self-pacing, and accessibility across geography.

The design logic changes completely

• In-person: Design for energy, discussion and relationship. Your job is to create conditions for insight.
• Online: Design for attention and clarity. Make every minute feel worth staying for.

The hybrid trap

Most organisations default to hybrid not because it's the right choice, but because it's the compromise. A well-designed online module and a well-designed in-person session are almost always better than a mediocre hybrid of both.

The biggest limitation of traditional eLearning is passivity. The solution isn't more animation — it's making the learning tool-based, not content-based.

What embedded tools look like in practice

• A risk assessment calculator inside a safeguarding module
• A decision tree walking frontline workers through a real referral process
• An interactive case study where choices have consequences and feedback
• A reflection journal that generates a personalised summary at the end

Why it matters for social sector organisations

When training workers on complex topics — family violence, trauma, cultural safety — passive learning is a missed opportunity. Interactive tools force application, not just recognition.

Most change management fails not because the strategy is wrong, but because people can't see where they're going. Design is the act of making the invisible visible — and in change processes, that is everything.

What design can do in a change process

• Journey maps — show people where they are now, where they're going, and what the milestones look like
• Visual frameworks — make abstract strategy concrete and discussable
• Co-design sessions — when teams help build the visual, they own the outcome
• Communication templates — give managers consistent tools to cascade the message

When teams can see the change clearly, they can respond to it. That visibility turns resistance into participation.

SMEs are the most valuable — and most challenging — people in any learning design project. They will almost always want to include more content than your module can hold.

The core tension

SMEs optimise for completeness. Learning designers optimise for behaviour change. These are not the same thing — and the project will suffer if this tension isn't named early.

What I do from day one

• Align on outcomes: "What do you want people to DO differently after this training?"
• Show, don't tell: Show two versions and ask which one they'd finish as a learner
• Create a parking lot: Important info that doesn't fit goes into a resource library, not the bin
• Review cycles with clear rules: Review rounds are for accuracy, not redesign

The right question isn't "how do we justify the training spend?" It's "what is the cost of not investing in it properly?"

The hidden costs of bad training

• Staff turnover: Replacing one employee costs 50–200% of their annual salary when you factor in recruitment, onboarding and lost institutional knowledge
• Compliance risk: In family violence, child protection and multicultural services, undertrained staff create real legal and funding risk
• Community harm: Workers make avoidable mistakes with the most vulnerable people your organisation serves
• Repeated training cycles: When training doesn't stick, organisations run it again and again

What to tell your board

Frame training investment in the language of risk mitigation and workforce sustainability. Help them see undertrained staff as an organisational risk, not a training line item.

Most nonprofits write their annual report for their auditor. The best ones write it for their next major donor. The difference in outcome is enormous.

What most annual reports get wrong

• They lead with financials instead of impact
• They use sector jargon that means nothing to external stakeholders
• They bury the most compelling stories in blocks of dense text
• They look like they were designed in Word at 11pm before the deadline

The design is the message

When a report is visually beautiful and clearly structured, readers assume the organisation is well-run. Your design communicates your capability before anyone reads a single word.

You spent three months recruiting the right person. Then gave them a two-day induction and a folder of PDFs.

What bad onboarding looks like

• Information dump in the first week with no time to process
• Compliance training that takes all day and is immediately forgotten
• No structured check-ins in the first 90 days
• Culture, values and ways of working explained verbally once and never reinforced

What good onboarding looks like

A well-designed onboarding journey is staged over 90 days. It introduces role, culture, systems and community context in a sequence that matches how people actually absorb information — with check-in points and space for questions, not just a box to tick.

A single 45-minute module delivered once a year does not change how workers respond to people from different cultural backgrounds in high-pressure moments.

What actually changes practice

• Embedded scenarios: Realistic situations that require decisions, not just information absorption
• Reflective practice tools: Structured ways for workers to review their own interactions and identify growth areas
• Leadership modelling: Leaders who visibly practise cultural safety signal it's a real organisational value
• Community voice in design: Training designed with input from the communities it aims to serve is always more effective than training designed about them

The most common metric organisations use to evaluate training is completion rate. But did anything actually change?

The Kirkpatrick Model — simplified

1. Reaction: Did learners find it relevant and engaging?
2. Learning: Did they actually acquire new knowledge or skills?
3. Behaviour: Did their behaviour change back on the job?
4. Results: Did organisational outcomes improve as a result?

Most organisations only measure level 1. The real value is in levels 3 and 4 — and those require deliberate design from the start.

Practical steps

• Define what success looks like in observable behaviour before you build the training
• Build in 30/60/90 day check-ins that ask managers what they've noticed
• Use scenario-based assessments that simulate real-world application
• Track the problem the training was designed to solve — not just whether people did the training