Hello friends,

Welcome back to Theory of Change, where I try to help leaders, founders, and creators build organisations that donors trust, teams love, and society actually needs.

Recently, we’ve been unpacking ideas borrowed from the marketing and tech world that tend to fall apart when they meet the messy realities of nonprofit life.

This week: User Personas.

(New here? Maybe subscribe?)

WHAT THEY'RE SUPPOSED TO BE

Many of you know that I’m a recovering Chief Product Officer. Back in the mists of time, when TikTok was just a sparkle in GX Capital’s eye, I led five teams of developers and designers stewarding about twenty digital tools and products.

In those multidisciplinary teams, I always saw the Product Owner as the voice of the user inside the room, an advocate who keeps the engineers honest and the designers grounded. Their job was to channel real human experience into technical decisions: to make sure that every feature, release, or sprint led back to someone’s actual need.

And that’s where user personas come in. A user persona is a fictional, evidence-based character that represents a segment of your audience or service users. It combines data, observations, and a bit of storytelling to help teams visualise who they’re designing for and what those people might need. Done properly, they’re meant to make the invisible visible, turning anonymous data into something the team can relate to.

Maria, 37, recently arrived in the UK. She’s family-oriented, pragmatic, and values self-reliance. She struggles with online forms but wants to feel independent when dealing with official services.

The persona acts as a proxy for empathy, a shared shorthand that helps people remember who they’re building for and why.

But I've never liked them.

WHERE THEY FALL APART

Three big problems:

1. They individualise systemic issues.

Personas often compress structural barriers into personality quirks. “Maria, 37, practical and independent, struggles with digital forms” sounds like insight, but it hides the real causes. Maybe her immigration centre was defunded. Perhaps the forms are only in English. What if the issue is that she’s using a shared phone with patchy data? A persona won't often pick these up.

2. They fake empathy.

Personas can simulate closeness without doing the work. We imagine what people might feel, but don't spend enough time and energy checking if we’re right. The result is a convenient stand-in for relationships we never actually properly build.

3. They freeze people in time.

Personas are two-dimensional when our work needs four dimensions: context, time, relationship, power. Real lives shift daily and our work is about systems and people in motion. Cardboard cut-out personas (replete with stock photography or ChatGPT-generated portraits) just don’t, er… cut it.

A BETTER WAY TO THINK ABOUT IT

Luckily, there are more dynamic, relational ways to understand people and systems. There are alternative approaches. For instance, I touched upon the benefits (but also limitations) of "Jobs To Be Done" in #052. It's better but not perfect.

Ultimately, thinking differently here rests on two shifts:

1. From marketing profile, to networked actor

Instead of “Maria, 37, values trust and convenience”, start with thinking about who Maria is connected to. What roles does she play? Where’s the leverage?

Community organisers use power-mapping for this. Sketch the system around a problem, ask who holds influence, who’s affected, who connects them.

So, for instance, if you are trying to understand extractive rental systems, you don’t design for “frustrated renters”. You map landlords, council officers, contractors, local media, and tenant networks and then build power where relationships already existed. Read on for tools that help you do this.

B. From empathy placeholder, to actual dialogue

Once you’ve mapped collaboration across the whole ecosystem - who works together, who’s isolated, how information travels - the next challenge is how to keep this real, and up to date. You need to keep empathy alive through continuous feedback grounded in experience.

That means building lightweight, ongoing conversations with the people inside that map, not as archetypal users (*shudders*), but as peers.

And that doesn’t need to be a grand engagement strategy. It can be a rhythm. A ten-minute check-in, a quick voice note, a shared doc where people add what’s changing in their world. The aim isn’t to extract new data every time; it’s to stay in a conversation.

TRY IT THIS WEEK

Sketch a quick system map.

Download Mobilisation Lab’s System Map Template and spend an hour mapping what actually upholds the problem you’re trying to change: the relationships, policies, habits, and incentives that keep it in place. Add arrows for influence, tension, or support. You don’t need perfect data; the point is to make the invisible visible.

Research constraining and enabling factors

Use Oxfam’s extremely comprehensive guide to influence and impact. The whole thing is worth a read, but skip to page 47 if you want to get straight into how to research influence structures.

There’s also a nice 2023 case study (related to UK housing) here, that I’ve found useful to see how fruitful this whole process can be, end-to-end.

And, as ever, there's no substitute for getting people into a space IRL. I've always been lucky enough to work alongside people who are amazing at getting people together in a way that is both relaxed and productive. If you have one of these unicorns in your team, hold on to them. Let them do their thing.

🌊 WAVE GOODBYE 🌊

If you’re leading a small organisation, I know what you’re thinking: another email, another damn mapping exercise? I don't have time!

But slowing down to see the system, slowing down to understand who’s connected, who’s excluded, and how power actually moves isn’t a time sink, it’s a time saver. Every hour you spend understanding the web around your work saves you ten spent fixing misunderstandings, chasing the wrong partners, or designing for ghosts.

It's something I've not been doing enough of recently. I talk about why (and what I plan to do about it) in my monthly, more personal studio newsletter.

If any of my free databases, courses, frameworks, GPTs, newsletters, or videos have helped you out, consider tipping me so I can continue to make this work. The generosity of some kind souls this month really blew me away.

Thanks for being here,

Adam