Data roadmap: the method for winning your executive committee over

Plenty of organizations have data initiatives. Few have a roadmap their leadership signs off and follows over time. The symptom is always the same: a data department that has invested in a platform, in tools, sometimes in AI, without ever finding traction against its executive committee's business priorities.

Nicolas Aumaistre (Head of Consulting, KPC) and Laurent Dresse (Global Head of Partner Success, DataGalaxy) shared a 3-phase method during a joint webinar KPC × DataGalaxy on building data roadmaps, on 16 June 2026: 45 minutes online, with a live demo of Portfolio Manager. The running theme: start from the business need, not the technology, and replace the PowerPoint document with continuous steering.

Why most data roadmaps fail

The pattern keeps repeating. An organization decides to put a data platform in place. The technical component goes in first. Then comes the question of which use case to put on it: the data foundation gets built without matching business needs, because the governance principles were never discussed up front.

The result: a gap between what the business needs and the infrastructure. Finance or operations looks at the investment and cannot see where the value is. The project stays at POC stage.

That picture matches a broader reality about AI in companies: most organizations already use AI in some business function, but in isolation, with no overall view and no dedicated governance. That is why so many AI projects fail to deliver the business value they promised. The real issue is slightly different: AI is not sitting on a foundation of data that is reliable, steerable, understandable and under control. So nobody controls how it is used, and it behaves like a black box.

What 3-phase method lets the business steer?

The approach presented runs in three stages.

• Phase 1: the exploratory audit. Understand the environment, by asking the executive team directly about their ambitions over 2, 3 or 5 years. A company aiming for growth through acquisition does not have the same data priorities as one working on its margin in a mature market. The audit covers 4 axes: the business stakes, the organization and governance in place, business uses and needs, and the existing technical architecture.

• Phase 2: strategic alignment. From the gaps the audit reveals, you define a target architecture and, above all, spot the quick opportunities: the use cases that will produce a visible result in the short term and act as leverage for winning the executive committee over.

• Phase 3: the transformation plan. You rank the workstreams, estimate their value and plan the milestones. And you steer the budget continuously rather than freezing a 3-year roadmap.

One principle runs through all three phases: the use case is never chosen at random. Eight criteria qualify it:

• A link to company strategy

• Visibility for the executive committee

• A real business pain point

• A committed business sponsor

• A limited scope

• A result achievable within 3 months

• An implicit improvement in governance

• Technical dependencies under control

A use case that ticks most of those criteria becomes the first milestone, the one that gets the executive committee's green light and pulls the rest along.

A concrete case: growth through acquisition and scattered data

A major healthcare player illustrates how it works. Its stated ambition: bring in two or three new entities within 3 years. That fact alone shapes the whole diagnosis. Growth by acquisition means, in almost every case, different ERPs from one entity to the next. So heterogeneous transactional foundations to consolidate in order to steer the company as a whole.

The audit revealed a siloed organization, data roles already named (data owners, data stewards) but poorly equipped for their actual responsibilities, and a recently created data department whose place in the hierarchy limited its influence on the executive committee. The existing use cases were mostly one-off BI projects, to get out of Excel spreadsheets. A partial answer, with no overall view.

The work followed the three phases:

• Understand the landscape: executive committee interviews, mapping the data roles and the organizational model, an inventory of the use cases, mapping the BI estate and the data sources.

• Adjust the course: mapping maturity, an inventory of the priority business workstreams, target architecture.

• Build the roadmap: a phasing strategy, macro-planning and budget, a summary presentation to the executive committee.

Two functions were chosen: finance, the company's backbone, and logistics, the operational heart of the business. The first use case consolidated the financial data from both existing ERPs with the inventory management tools, cross-referencing financial data with stock valuation data. A deliberately tight scope, with no ambition to migrate everything at once onto a centralized platform.

The principle is clear: recreating the heterogeneity of the source systems inside a new platform solves nothing, it just moves the problem. Building a use case is also the chance to clean and harmonize the data at source. Fourteen months on, the roadmap is still rolling out and widening its scope, with AI topics gradually being brought in.

Why steer rather than present?

A data roadmap presented top-down usually gets surface agreement. Everyone nods, nobody really acts. The difficulty shows up afterwards: linking every euro invested to measurable business value, at the moment the executive committee or finance starts questioning infrastructure and governance costs.

That is where the second dimension of the method comes in: tooling the steering. Most organizations today manage their use cases in an Excel file or equivalent. Flexible locally, but neither governed nor centralized across the company.

Portfolio Manager, DataGalaxy's add-on built for this, works as a central register of use cases: every initiative is linked to a company priority, a status, an affected business unit and an estimated value. The tool does not replace the audit or the method. It makes visible, in real time and with the right security levels, the correlation between strategic ambition and how the workstreams are actually progressing.

The roadmap then becomes a cockpit: when an external factor changes (an acquisition, a different economic context), the use case priorities can be readjusted without starting from a blank page.

What it changes concretely

• Diagnose before you prioritize. A structured audit across the 4 axes (business, organization, uses, architecture) always comes before choosing the workstreams. Without that diagnosis, the roadmap answers a technical hunch, not a documented business need.

• Choose a first use case that ticks the right criteria. A real business pain point, a committed sponsor, a limited scope, a result within 3 months. That first success is what funds and legitimizes the ones after it.

• Steer continuously, rather than presenting once a year. The roadmap is not a frozen deliverable. It is a living thing, where every workstream stays linked to a value that can be measured and adjusted over time.

From the method to your own context

Building a data roadmap is not a solitary exercise or a project: it is a state of mind and a genuine way of managing data initiatives day to day. It rests on understanding what is at stake for the business, in data and organizational terms, a fast, iterative delivery model, a 360° view (architecture, tooling, organization, governance), a culture of ROI, communicating the roadmap at every level and strong sponsorship from management.

The method applies to companies of very different sizes and maturities, broader or narrower, deeper or lighter depending on the ambition. What does not change: start from the business need, choose a first use case that produces a visible win, and instrument the steering rather than presenting it once.

Data macro-framing workshop: free

KPC offers a free 2-hour data macro-framing workshop to apply this diagnosis to your organization: identify your priority use cases and lay the groundwork for your roadmap. The workshop bridges KPC's consulting approach and DataGalaxy's Portfolio Manager tooling for steering it over time.