Data storytelling: turning a chart into a decision

What decision should this chart change? If you can't answer that before building it, the chart probably won't change anything. Data storytelling is the practice of presenting data so that a specific audience understands one clear point and can act on it. It's less about narrative flourish and more about choosing what to show, what to leave out and what to say plainly.

In a senior meeting, the best data story is often a single chart and three sentences.

Why start with the decision?

Because it sets everything else. If the decision is whether to keep a second facing, the story is about sell-outs and sales per facing. If it's whether to repeat a promotion, the story is about incremental sales and the post-promotion dip. Different decisions call for different evidence.

Starting with the decision also makes it easier to cut. Most analyses produce far more charts than a meeting can absorb. Knowing the decision lets you keep the two that matter and move the rest to an appendix.

And it respects the audience. Senior readers want to know what's being asked of them. Stating the decision up front, then showing the evidence, gets to the point faster than building up to it.

How do you make one chart carry one point?

Choose the right form. A line for change over time, a sorted bar for comparisons, a scatter for relationships. Good data visualization practice favors simple forms that most readers already know how to read.

Highlight what matters. If the point is that one store group diverged from the rest, color that group and gray the others. If the point is a change after a specific date, mark the date.

Write the takeaway as the title. "Test stores pulled ahead after the display went in" tells the reader what to see. "Weekly units by store group" makes them work it out.

What context should accompany the chart?

Enough to prevent misreading. The period, the store basis, the definition of the measure. Any unusual event in the period, such as a holiday shift or a change in store count.

The comparison. A number without a baseline is hard to judge. Show last year, the control group or the category, whichever fits the decision.

The uncertainty. If the result is a range, show it. If the sample is small, say so. Senior readers are generally comfortable with uncertainty when it's stated clearly. They're less comfortable discovering it later.

It also helps to place the result in the wider market. A category's performance in the monthly NRS Insights reports can show whether your result reflects a broader trend.

How should the story be presented?

Briefly. State the decision, show the chart, give the takeaway, note the main caveat and make the recommendation. That sequence works in a meeting, in an email and in a written report.

Invite challenge. A good data story holds up to questions. Having the supporting detail ready, and welcoming questions about it, builds more trust than a polished presentation that can't be probed.

And follow up. After the decision, report what happened. Over time, a record of recommendations and their outcomes is the strongest story an analytics team can tell. The July 2026 same-store sales report is an example of monthly context that can sit behind those conversations.