From Data to Decision: Turning Numbers Into Analytical Judgement

There is a comfortable myth that data makes decisions. It does not. Data informs decisions. People make them. And the gap between having data and making a good decision is wider than most organisations admit. It is entirely possible to be data-rich and decision-poor, to have dashboards full of metrics and still make choices that the data does not support, because nobody did the work of turning the numbers into judgement.

That work is the Business Analyst's contribution. Reading data and interpreting it honestly, which we covered on the previous two days, are the foundations. Today is about the structure that sits on top of them: how to move from an honest interpretation to a recommendation that a decision-maker can act on with confidence.

Data Informs, Judgement Decides

The first thing to be clear about is the relationship between data and decision. Data is an input. It is evidence. It reduces uncertainty, but it rarely eliminates it. Almost every real business decision involves factors that the data does not capture: strategic priorities, risk appetite, the political context, the things that are known but not measured. This means that a good decision is not simply the one the data points to. It is the one that weighs the data appropriately alongside everything else that matters. The BA who understands this does not present data as if it settles the question. They present it as the strongest available evidence, clearly interpreted, with its limitations stated, as one input into a judgement that belongs to the decision-maker.

This framing matters because it positions the BA correctly. The BA is not there to make the decision or to use data to force a particular outcome. The BA is there to ensure the decision-maker has the clearest possible understanding of what the evidence shows, so that the judgement they make is as well-informed as it can be.

Building a Recommendation the Data Supports

A recommendation built on data has a structure. Working through that structure is what separates a genuine evidence-based recommendation from a conclusion that has been decorated with a few supporting numbers.

1. Start with the question. What decision is this recommendation supporting? A recommendation that is not anchored to a specific decision tends to drift into a general presentation of data that leaves the decision-maker no better placed to choose. Be explicit about the question the recommendation answers.
2. Then state what the data shows. Not everything in the dataset, but the findings that are relevant to the question. Present them in plain language, honestly interpreted, using the principles from the previous two days. This is the evidence.
3. Then state what the data does not show. Every recommendation has limitations, and naming them is what makes the recommendation trustworthy. What factors are relevant to this decision that the data does not capture? What assumptions does the interpretation rest on? Where is the evidence weaker than you would like? A decision-maker who knows the limitations can weigh the recommendation appropriately.
4. Then give the recommendation. Based on the evidence and its limitations, what do you recommend, and why? The recommendation should follow logically from what came before. If a reader can trace the line from the question, through the evidence, through the limitations, to the recommendation, then the recommendation is well-built. If the recommendation appears from nowhere or requires the reader to take a leap that the evidence does not support, it is not.

Presenting Data So It Informs Rather Than Manipulates

Yesterday we covered the traps that make data mislead. Today the responsibility flips. When you present data, you have the same tools available that misleaders use, and the discipline is to use them honestly.

• Start your axes at zero unless there is a genuine analytical reason not to, and if there is, say so.
• Choose time periods that are representative rather than flattering. Present averages alongside the spread when the spread matters.
• Give raw numbers their denominators.
• Show the data that complicates your recommendation as well as the data that supports it.

The principle underneath all of these is simple. Present data the way you would want it presented to you if you were the one making the decision and the outcome mattered to you personally. The honest presentation is the one that gives the decision-maker the clearest possible view of reality, not the one that nudges them toward the conclusion you prefer.

There is also a clarity dimension to this. Data presented in an overly complex way fails to inform even when it is honest. A decision-maker who cannot understand the chart cannot use it. The skill of presenting data well includes the skill of simplifying it to the point where its meaning is clear, without simplifying it to the point where it becomes misleading. That balance is a craft, and it is one the best BAs develop deliberately.

When the Data and the Stakeholders Disagree

One of the most difficult moments in data work is when the honest interpretation of the data conflicts with what stakeholders want, expect, or believe. This happens regularly. A sponsor is committed to a course of action and the data does not support it. A team believes their initiative succeeded and the data is ambiguous. An organisation has invested in a direction and the data suggests it is not working. In these moments, the BA is caught between the integrity of the interpretation and the pressure to give people the answer they want. The way through is not to soften the interpretation until it becomes acceptable. That destroys the value of the BA as an honest broker. Nor is it to deliver the inconvenient finding bluntly and let the relationships take the damage. The way through is to deliver the honest interpretation with clarity and with care.

Clarity means being unambiguous about what the data shows, not hiding the inconvenient finding inside qualifications until it disappears. Care means delivering it in a way that respects the stakeholder, acknowledges what they were hoping for, and frames the finding as useful information rather than a verdict. The data does not support the conclusion we hoped for is a sentence that can be delivered in a way that builds trust or destroys it. The difference is in the care. It also helps to separate the finding from the decision. The BA's job is to ensure the interpretation is honest. The decision still belongs to the decision-maker, who may choose to weigh other factors more heavily than the data. By being clear that you are providing the evidence rather than dictating the choice, you can deliver an inconvenient finding without it becoming a confrontation. You are not telling the sponsor they are wrong. You are giving them the clearest possible picture so that their decision is well-informed.

The BA as the Bridge

The movement from data to decision is the point in the analytical process where the Business Analyst adds the most distinctive value. Data analysts can produce the numbers. Decision-makers can make the call. The BA is the bridge between them, the person who turns evidence into a form that a decision-maker can actually use, honestly interpreted and clearly presented. This is why data competence matters so much for the BA, and why it is worth the effort of building it. The BA who can read data, interpret it honestly, and turn it into a clear recommendation is operating at the centre of how modern organisations make decisions. That is a position of genuine influence, and it is available to any Business Analyst willing to build the skills this week has covered.

Go out and be successful.

Oluwatosin Ogunkoya | Flotog BA Insights | www.flotogbainsights.com

Tomorrow: The Data Analysis Toolkit. A practical reference guide to everything in this series, with a free downloadable resource to use on your next data-informed project.