The Data Strategy That Works — A 6-Step Framework (Part III)

The Data Strategy that Works [Petgrave.io]

In Part I, we talked about understanding your business and knowing what to measure. In Part II, we asked two harder questions — whether your organisation is actually capable of using data, and where data breaks down before it ever reaches a decision.

If you have worked through all of that honestly, you are in a position most organisations never reach. You know how your business creates value. You know what success looks like. You know your real capability. You know where the gaps are.

Now comes the part where most organisations make their final and most costly mistake.

They take everything they have learned and immediately start building. A dashboard here. A new tool there. An AI pilot someone read about in a newsletter. Activity without direction. Solutions without a clear sense of what problem they are actually solving — or for whom.

Part III is about avoiding that mistake. It covers two things: deciding what is genuinely worth building, and being clear about where your data strategy is actually heading.

Step Five: Deciding What Is Worth Building

Every organisation has more data problems than it has the capacity to fix. Every team has more ideas than it has the time or resources to execute. The question is never simply what should we build? It is what is actually worth building — and for whom?

This is the discipline most data conversations skip entirely. People talk about tools, platforms, dashboards, and AI before they have answered the most fundamental question: if we build this, who will feel the difference — and how?

If that question cannot be answered clearly, the initiative should not move forward. That is not pessimism. It is the filter that separates data strategies that deliver from ones that just consume resources.

Here is how to apply it.

• Start with the person, not the problem

Before you describe any solution, you need to deeply understand the person it is meant to serve. And in data work, that person is often internal — an operations manager, a team lead, a founder, a service coordinator. They deserve the same rigour you would apply to understanding an external customer.

Ask yourself: what is this person actually responsible for achieving? What does their success look like — not on paper, but in the reality of their working day? What are they judged on? And where does failure hurt them most?

Go beyond the functional. Consider how they want to feel. A manager who has to make resource decisions every Monday morning does not just want data — they want to feel confident that the decision they are making is the right one. That emotional dimension matters, because if the solution you build does not address it, people will not use it regardless of how technically sound it is.

• Get specific about the pain

Once you understand the person, get ruthlessly specific about what makes their job difficult today. Not in general terms. In the actual, lived detail of their work.

Ask yourself: where is time being wasted right now? What creates uncertainty or delay in their decision-making? What data is missing, or late, or simply not trusted? What risks keep recurring because there is no reliable way to see them coming?

The specificity matters. Vague problems produce vague solutions. If someone says "we don't have good visibility," that is not specific enough to build anything useful around. Press further: visibility into what? For what decision? At what frequency? By the time you have answered those questions, you are much closer to something worth building.

• Define what success looks like from their perspective

Before thinking about what to build, describe what the person would experience if the problem were solved. Not what the system would do — what they would be able to do differently.

Ask yourself: what decisions would become faster or more confident? What manual effort would disappear? What uncertainty would be removed? What would they be able to do that they simply cannot do today?

This is the value you are trying to create. Everything else — the technology, the platform, the process — is just the means of getting there. If you cannot describe the value in terms the person would recognise and care about, you have not found it yet.

• Name the initiative in outcome terms, not technology terms

This is a simple but revealing test. When you describe the thing you want to build, do you reach for what it is or what it delivers?

"An AI dashboard" describes a technology. "A daily view that tells operations managers where demand is building before it becomes a problem" describes an outcome. The second one is worth building. The first one might be, but you cannot tell yet.

Ask yourself: can you describe the full value of this initiative without mentioning any technology at all? If the answer is no — if the justification for building something is essentially "because AI" or "because data" — pause. The technology is not the reason. The value is the reason. Find the value first.

• Apply the filter honestly

Before any initiative moves forward, run it through these questions:

Would the person this is built for actually notice if it did not exist? If the honest answer is probably not, question whether it is worth building at all.

Does this fit what your organisation is genuinely capable of right now? There is no point building a Level 4 solution for a Level 2 organisation. The foundation has to support the structure.

Is the solution proportional to the problem? Not everything needs to be automated. Not every insight needs a platform. Some problems are best solved by a clear process, a shared document, or a well-run weekly conversation. Use the simplest solution that actually works.

Is there someone who will be accountable for acting on what this produces? Data that nobody acts on is not an asset. It is overhead. Every initiative needs a clear owner — not for the technology, but for the decision it is meant to support.

These questions will eliminate some ideas you were excited about. That is exactly what they are supposed to do. The organisations that get data right are not the ones that build the most things. They are the ones that build the right things — and say no clearly to everything else.

Step six: Knowing Where You Are Heading

Once you know what is worth building, there is one more question that has to be answered — and it is the one that holds everything together.

Where is this all going?

Not eventually. Not in five years when everything is perfectly optimised. Just the next meaningful step. The direction that gives every data decision a reference point. The statement that lets people say yes to some initiatives and no to others, with confidence and without endless debate.

That is what a data vision is. And most organisations either do not have one, or have one so vague it provides no actual guidance.

Here is what it needs to contain.

• Be honest about where you are heading next — not where you want to be eventually

A data vision is not a destination. It is a direction. And the most important word in it is next — not ultimately, not one day, not when we are ready.

Ask yourself: what is the next realistic step forward for this organisation, given where we genuinely are today? Which capability needs to improve first, because everything else depends on it? What would meaningfully better look like in the next twelve to twenty-four months — not as a fantasy, but as something we could actually achieve?

The organisations that set realistic directions build credible momentum. The ones that set grand visions with no connection to current reality create cynicism — because the gap between the vision and the daily reality becomes demoralising rather than motivating.

• Describe how decisions will change — not what technology you will use

This is the heart of any honest data vision. Forget the tools. Forget the platforms. Describe the human change.

Ask yourself: which decisions will be made differently because of this strategy? Who will make them with more confidence, more speed, or more evidence than they do today? Which decisions currently made on instinct will move toward being grounded in data? Who benefits most from that shift — and how will their working life be different?

Be specific. A vision that says "we will be more data-driven" is not a vision. A vision that says "our operations team will make resourcing decisions based on real demand signals rather than last month's averages, so they can respond in hours rather than weeks" — that is a vision. It describes a specific decision, made differently, by specific people, with a tangible outcome.

• Connect the direction to business value

The decision change has to lead somewhere the business actually cares about. Not just faster reports or cleaner dashboards — real outcomes. Reduced costs. Better service. Lower risk. More confident growth decisions. Revenue that does not leak.

Ask yourself: what business outcomes improve because decisions get better? What risks reduce? What becomes possible that was not possible before? What effort disappears that currently consumes time and energy without proportionate return?

If the connection between the data direction and the business outcome is unclear, strengthen it before you share the vision with anyone. Leaders will always ask: what is this for? The answer needs to be immediate and concrete.

• Be explicit about what you will not do

This is the most underrated part of any data vision — and the part most organisations leave out entirely.

A direction without boundaries is not a direction. It is a wish. And wishes invite scope creep, hype-driven detours, and the kind of unfocused activity that leaves organisations busy but not progressing.

Ask yourself: what is explicitly out of scope right now? Where will you not use automation or AI yet — because the foundation is not ready? What will remain human-led by design, because judgment matters more than speed in those decisions? What are you choosing not to pursue, even if it sounds exciting?

These boundaries are not weaknesses. They are signals of maturity. They tell everyone in the organisation — and anyone outside it — that the strategy is grounded in reality, not trend-chasing.

• Test it before you commit to it

Before a data vision is shared, run it through a few honest checks.

Could someone with no technical background read this and understand what it means and why it matters? If it requires specialist knowledge to parse, it is not clear enough yet.

Does it feel achievable given where the organisation genuinely is today? If it sounds like a vision for an organisation three times more mature than yours, scale it back. An honest, achievable vision that is delivered builds more trust than an ambitious one that is quietly abandoned.

Would this help someone say no to a bad idea? A good vision is a filter. When someone proposes an initiative that does not fit, the vision should make it obvious — without needing a long meeting to work that out.

Does it connect clearly to something the business cares about? If the link to real outcomes is not immediately visible, it will not get the commitment it needs to succeed.

The Complete Picture

Across three posts, we have worked through the full arc of thinking that a data strategy has to be built on.

Understanding how the business actually creates value — not as it appears on a strategy slide, but as it works in practice every day. Knowing what to measure, and having the discipline to track only what informs decisions. Being honest about what the organisation is genuinely capable of right now. Seeing clearly where data breaks down before it becomes useful. Deciding what is actually worth building, and for whom. And pointing clearly in a direction that is grounded, bounded, and connected to outcomes that matter.

None of this requires the most sophisticated technology. None of it requires a large data team or an expensive consultant. It requires clarity, honesty, and the willingness to ask uncomfortable questions before reaching for solutions.

That is what a data strategy that works actually looks like. Not a document filed away after a planning retreat. A living practice — the discipline of connecting information to decision-making, consistently and deliberately, in a way that makes the business better.

The organisations that get this right do not always start with the best tools. They start with the clearest thinking.

---

Petgrave.io helps founders, teams, and organisations build data strategies grounded in business reality. If this series raised questions worth exploring in your organisation, we would be glad to talk.