Torras.ai

Welcome to the Torras AI Podcast. In this episode, I speak with Ignasi Alcalde, a consultant, university lecturer, and author who specialises in data visualisation, data storytelling, data literacy, and generative AI. We explore a deceptively simple question: if organisations have more data, more dashboards, and more artificial intelligence than ever, why do so many decisions remain difficult—or fail to improve at all?

The conversation is in Catalan. A summary of the conversation appears below. Ignasi's central message is clear: data has value only when it helps someone understand a situation, exercise judgment, and make a better decision.

Watch the conversation

Highlights from the conversation

1. More data does not automatically produce better decisions

Ignasi's work began with information visualisation: how to transform data into charts and diagrams that people can understand. Over time, however, he saw that a good chart was not enough. An attractive visualisation can still fail if nobody knows what it means, why it matters, or what to do next. That led him from visualisation to data storytelling—and from storytelling to the broader problem of decision-making.

Many organisations already have sophisticated analytics platforms, Power BI dashboards, and increasingly powerful AI systems. Yet when Ignasi enters a company, his first question is not about its technology. He asks: “What decision do you want to make with these data?” If the answer is unclear, the real problem is usually not a lack of tools. It is a lack of good questions, shared purpose, and decision culture.

2. Start with the decision, not the dashboard

The most useful sequence is the reverse of what many teams do. Do not begin with all the available data and hope that a decision eventually emerges. Begin with the decision, formulate the question, identify the evidence needed, perform the analysis, and then communicate the result. In compact form: decision → question → evidence → analysis → communication.

This order creates focus. It also reveals which information matters, which assumptions are being made, and what evidence could change someone's mind. A relevant insight is not merely an interesting pattern in a dataset; it is a finding capable of changing a strategic decision.

3. The silent failure of dashboards

Dashboards rarely announce that they have failed. They continue to refresh, accumulate indicators, and appear in meetings. Ignasi proposes a sharp diagnostic question: “If we removed this dashboard tomorrow, which decision would become worse?” If nobody can name one, the dashboard may be reporting activity without creating value.

A dashboard should not try to prove everything an organisation knows. Its purpose is to make a decision easier. A useful structure starts with a small set of essential indicators, offers a diagnostic layer for understanding why something is happening, and ends with an action or outcome layer. When every possible metric is placed on the same screen, attention is diluted and the important signal disappears.

4. Executives need answers, not additional noise

At an executive level, the essential questions are remarkably stable: What is happening? Why is it happening? What can we do about it? Better analytics should help answer those questions, not simply create more reports. This requires a combination of analytical rigour and narrative clarity: evidence must be accurate, but it must also be presented in a form that makes its implications understandable.

The cultural challenge appears when the evidence contradicts intuition. Instead of revising an opinion, people may search for another metric that confirms what they already believe. Data literacy therefore involves more than knowing how to read a chart. It also requires the discipline to question assumptions and accept inconvenient results.

5. AI changes the analyst from producer to orchestrator

Generative AI can now clean data, propose analyses, create first drafts of visualisations, and suggest possible narratives in minutes. That does not eliminate the need for professionals; it changes where their attention is most valuable. The analyst's role moves away from producing every intermediate artefact and towards orchestrating the process: framing the problem, checking sources, validating calculations, comparing hypotheses, interpreting results, and deciding what deserves to be communicated.

Ignasi calls the final step last-mile analytics: turning a technically correct result into a clear narrative that can actually move a decision. AI can accelerate the earlier stages, but responsibility for the final judgment remains human.

6. AI is most useful as a demanding thinking partner

While writing Data Storytelling with Artificial Intelligence: From Data to Decisions, Ignasi used AI openly for roughly 10–20% of the work. It helped him explore ideas, test the book's structure, surface counterarguments, and simplify explanations. He did not delegate the thesis, personal judgment, or point of view.

That distinction matters. AI can be a copilot or sparring partner that introduces productive friction: challenge this argument, show me a weakness, or offer a plausible alternative. The danger begins when we use it to avoid thinking altogether. As Ignasi puts it, there is a profound difference between using AI after thinking and using AI instead of thinking.

7. Data literacy does not require becoming a data scientist

Executives do not need to become programmers or statisticians, but they do need to ask better questions. Where did the data come from? Who is included or excluded from the sample? Are we confusing correlation with causation? Is a truncated axis exaggerating a difference? Which assumptions drive the result? Which parts were generated or inferred by AI?

Ignasi describes this as wearing three complementary hats: the analyst's concern for evidence, the designer's concern for clarity, and the journalist's instinct to investigate context and ask what may be missing. Together, these habits make us less vulnerable to confident but misleading presentations.

8. Ethical storytelling is about fairness, not the illusion of neutrality

Every act of communication involves selection. We choose a chart, a time range, a comparison, and an order in which to present the facts. Complete neutrality is therefore difficult to claim. The ethical boundary lies between helping an audience understand reality and steering it towards a conclusion that the evidence does not fairly support.

Ignasi offers a practical test: If the audience knew all the relevant information, would they still consider this presentation fair? It is a powerful question for anyone who communicates with data, whether the work was produced by a person, an AI system, or both.

9. The augmented professional protects judgment and critical thinking

The most valuable professionals will not simply produce faster. They will ask better questions, understand context, connect ideas that appear unrelated, and exercise judgment when the available evidence is incomplete. These capabilities become more important—not less—as AI makes basic production easier.

Critical thinking is also the capability most at risk of atrophy. If an AI always provides the first interpretation, the first structure, and the first answer, it becomes easy to stop forming our own view. The augmented professional uses the machine to extend thought while remaining responsible for the outcome.

10. One habit every company can adopt tomorrow

Ignasi's recommended first step requires no new software: begin meetings and projects by asking, “What decision are we trying to make?” The question clarifies what information is relevant, exposes hidden assumptions, identifies what could change the group's opinion, and connects the analysis to a concrete action. It also makes responsibility visible: somebody must ultimately own the decision.

Technology matters, but the path from data to decisions begins with purpose. A dashboard, a story, or an AI model is useful only when it helps people see more clearly and act more wisely.

Continue exploring Ignasi's work

You can learn more on Ignasi Alcalde's website, read his professional profile, or connect with him on LinkedIn. His book has dedicated websites in Spanish and English, and is also available from Amazon in Spanish and English. Readers can also explore the book's companion conversational assistant, which extends the material beyond the printed page.