In November last year we launched the Kindling programme and invited applications for £3,000 to support tech justice work. We received 92 applications, many of which were strong and worth backing.

We’re sharing what we’ve learned from the process. We’re doing this to offer something useful and insightful to everyone who applied, and to others designing justice‑aligned funding.

“There were loads of good projects. There’s loads of amazing work out there.” – Kate Swade, Director, Catalyst

Tech justice is not the same as tech for good

We tried to be clear that tech justice is not the same as “tech for good” or general digital capacity building. We were looking for work that names and tackles structural injustice in and through technology, not just useful tools or skills delivered to communities.​

This distinction kept coming up in the applications themselves, and in our conversations as reviewers – especially around how tech justice builds on (and is therefore slightly different from) traditional digital inclusion when it includes a clearer focus on power, marginalisation, and systemic imbalance. 

With that in mind, we invited proposals that could strengthen tech justice practice in a few different ways. Then the closure working group team (Ellie, Kate, Max and Joe), reviewed them.

4 reasons we said no, even to good ideas

1. “Good work” isn’t always “tech justice work”

A common theme in unsuccessful feedback was that proposals were justice‑oriented, important, and community‑rooted – but not tech justice as we had defined it. In many cases, technology was being proposed to further a justice‑oriented project, rather than being part of a project tackling a systemic, technology-derived injustice.

“A lot of applications were about justice-oriented projects, which are great and important… But not always the same as tech justice.” – Max Wayne, Assessor, Catalyst

2. The issue (and the analysis) wasn’t clear enough

Several proposals were turned down because they didn’t clearly define the tech‑justice issue, or didn’t show how the work would strengthen understanding of that issue. Sometimes the direction of travel also felt unclear: harms caused by tech were named, but the response moved quickly to “more tech” without enough explanation of why that was the right response in that context. Enough explanation could have shown how the new tech might be different/better, or evidence of research or co-design with the people most affected.

“It was unclear how this would essentially address any systemic imbalances and injustices.” – Ellie 

“I struggled where the framing seemed to be “Here are some problems caused by tech. The answer is more tech”. I needed some more analysis to explain why that was.” – Kate

3) Accountability didn’t show up on the page

We saw applications that talked about being community led, but failed to show how decisions would actually be shaped by those most affected. In contrast, the strongest applications made accountability legible by sharing details. They proactively and thoughtfully named the value exchange for participants so it was clear the process wouldn’t be extractive.

“Some applications used the words ‘community-owned’ or ‘working with the community’… but there wasn’t enough evidence of this in the rest of the application.” – Max

4. Scope didn’t match £3,000

Some proposals were simply too big for the grant size – especially where the plan implied building substantial new infrastructure or designing major new tech. Often we could see the value in the direction, but we couldn’t see a realistic slice of work (for example, a discovery phase) that could be held in a micro‑grant.

3 things our scoring told us about our own lenses

1. Some values are easier to “see” in an application

Across the scored applications, curiosity and equity tended to score highest, while love and interdependence scored lower on average. We treat this as a prompt for our reflection, not as a finding about the field of work.​

Max’s experience especially echoed how subjective love could be to score, and how much it depended on interpretation and conversation between reviewers.

“For me, love was the hardest to score because that’s such a subjective thing.” – Max

2. Scoring helped, but conversation mattered more

We experienced the scoring as useful when it supported fairness and consistency, while still leaving room for human judgement and challenge. This mattered because application forms can privilege people with time, confidence, and grant‑writing skill.​

“For me the scoring worked well because it was then part of a conversation rather than it just being scoring on its own.” – Kate

3. Criteria can create a target – and a performance

Our criteria helped give applicants a clearer target. At the same time, reviewers still needed judgement and conversation to avoid rewarding neat “criteria‑parroting” over authentic, situated proposals. That tension showed up strongly in how some applications repeated the language back to us without evidencing it.​

Note: we did see applications where values language felt shoehorned in, or where the writing felt auto‑generated and unedited. It’s important to name that we understand why people reach for shortcuts under pressure, and we also found it made it harder to trust the work behind the words.