Before you adopt another AI tool, make sure you know what you're building

The charity sector has embraced AI faster than almost anyone predicted. The next challenge is knowing what to do with it. 

The pace of change 

Something significant has happened in the charity sector over the past two years. 

In 2024, around 61% of charities were using AI tools in some form. In 2026, that figure has risen to 88% - almost nine in ten organisations, across every size and type, incorporating AI into their day-to-day work. The Charity Digital Skills Report 2026 documented the shift, and noted something equally striking: the gap in AI adoption between large and small charities has effectively closed. This is no longer a story about well-resourced organisations pulling ahead. It is a sector-wide shift, and it has happened quickly. 

That pace of change is worth acknowledging. The charity sector is not always credited with moving quickly on technology. The AI adoption data suggests a sector that is genuinely engaged with new tools, willing to experiment, and looking for ways to do more with the resources available. That is a meaningful shift in posture. 

The question now is what comes next. 

The gap between adoption and strategy 

The same report that recorded 88% day-to-day adoption found that only around 4% of charities are using AI strategically. That gap - between widespread use and considered application - is where the next challenge sits. 

But what does ungoverned AI adoption look like in practice? Across the conversations we have with the sector, a consistent pattern emerges. Individual teams adopt tools independently - a fundraising team using one platform, a communications team using another, a digital team experimenting with a third. Each is solving an immediate problem. None of them are working from a shared understanding of what the organisation is trying to build digitally, or how AI fits into that picture. 

The results are predictable. Effort gets duplicated. Outputs are inconsistent. Data sits in separate places, processed by separate tools, producing insights that can't easily be combined or compared. When a trustee or a senior leader asks what the organisation's AI strategy is, the honest answer is often: we don't have one. We have a collection of individual decisions that nobody has yet joined up. 

That's not a criticism of the people making those decisions. Adopting a tool that makes your work more efficient is a rational response to pressure. But at an organisational level, adoption without a plan produces scattered cost rather than compounding capability. And in a sector where every investment needs to be justifiable, that distinction matters. 

What strategic use actually looks like 

The charities getting the most from AI are not the ones with the most tools. They are the ones who knew what they were building before they started building it. 

In practice, strategic AI use tends to share a set of common characteristics. The data model is clean - information flows across systems in a way that makes it usable, rather than sitting in silos that AI tools can't connect. Governance is in place - there is a shared understanding of where AI is used, how outputs are reviewed, and what the boundaries are. And AI is treated as an extension of the digital strategy rather than a parallel track running alongside it. 

We have seen this work well in organisations we’ve worked with that took the time to understand their digital architecture before introducing new tools. The AI capability they built sits on a foundation that can support it - which means it produces consistent, reliable outputs that improve over time rather than generating noise that has to be manually filtered. The difference in what those organisations can do with the same tools, compared to organisations that adopted first and planned later, is significant. 

What that foundation requires is clarity about what the organisation is building digitally. Not a vague ambition - a specific, prioritised picture of what the digital presence needs to do, in what order, and what needs to be in place before the next layer can be added. AI is one of those layers. It works best when the layers underneath it are solid. 

The prerequisite many organisations skip 

There is a version of the AI conversation that starts with the tools - which platform to use, which teams should have access, what the policy should say. That conversation is worth having. But it is the wrong place to start. 

The organisations using AI well started somewhere different. They started with a clear picture of their digital direction - what they were trying to achieve, what was currently getting in the way, and what the right order of investment looked like. AI entered that picture as a natural next step, not as a response to sector pressure or a trustee asking why the organisation wasn't doing more with it. 

For organisations that have secured funding or have a strategy confirmed, the question is often not whether to invest in AI - it is how to make sure that investment lands in the right place, in the right order, as part of a coherent digital direction. That sequencing question is harder than it looks from the inside. The digital estate is complex, the tools are developing quickly, and the pressure to act can push decisions faster than the evidence warrants. 

Where to start 

For organisations that want to move from ad hoc adoption to strategic use, five things make the biggest difference: 

1. Understand what you already have.

Before introducing anything new, map the tools and platforms currently in use across the organisation. Not just what the digital team is using - what every team is using. The picture is often more fragmented than anyone realised, and identifying it is the first step toward knowing what needs to change. 

2. Establish where you are now and where you want to be.

An honest assessment of your current digital maturity - what's working, what isn't, and what the gaps are - gives you a baseline to build from. Without it, there's no reliable way to know whether a new tool is moving you in the right direction or adding to the complexity. 

3. Agree your digital direction before adding new tools.

AI works best as an extension of a clear digital strategy, not a substitute for having one. Knowing what you're trying to achieve digitally - and in what order - is what allows you to introduce tools that compound capability rather than duplicate effort. 

4. Treat your data as the foundation.

AI tools are only as useful as the data they work with. If your data is fragmented across systems that don't connect, the outputs will reflect that. Getting the data model right is less visible than adopting a new platform, but it produces more lasting results. 

5. Govern AI at an organisational level, not a team level.

Shared policies, shared understanding of where AI is used and how outputs are reviewed, and a clear picture of what the boundaries are - these are what turn individual efficiency gains into organisational capability. 

Getting that sequencing right is what Giant Direction is designed to help with. We work through your digital challenges with you - mapping what you have, identifying what's actually causing the results you're seeing, and delivering a prioritised roadmap of what to do now, next, and never. That roadmap is what turns a confirmed strategy into a delivery plan with AI embedded in the right places, rather than bolted on as an afterthought. 

About Giant Direction 

Giant Direction is a consulting service that gives organisations a clear, prioritised direction for their digital presence. If you have the ambition and the investment but aren't sure what to do first - or you've adopted tools without a clear picture of how they fit together - Giant Direction was designed for exactly this.