Structuring a Narrative Around Your Analysis

Start With the So-What

Before structuring a single slide or chart, write one sentence that states the conclusion and why it matters — the "so-what." If you can't state it in a sentence, the analysis likely isn't focused enough yet to present. This sentence becomes the spine the rest of the narrative hangs off of: every chart and finding you include should either support this sentence or provide necessary context for it, and anything that does neither is a candidate to cut.

A useful test: if someone only heard your opening sentence and then left the room, would they walk away with the correct, actionable takeaway? If the honest answer is no, the so-what isn't sharp enough yet.

Building a Logical Throughline

A narrative isn't just a sequence of charts in the order you happened to build them — it's an argument with a beginning, a middle, and an end. Start by establishing the situation and the question worth asking. Move through the evidence in an order that builds understanding, typically from the broadest pattern down to the specific detail that explains it, or from the most surprising finding to the context that makes sense of it. Close with what the evidence means and what should happen next.

Every transition between sections should answer an implicit "so what does that lead to?" If a viewer could reasonably ask "wait, why are we looking at this now?" the ordering needs work. A narrative that jumps between unrelated findings without connective logic reads as a report, not a story — technically complete, but forgettable.

Choosing What to Cut

Almost every analysis produces more findings than belong in the final narrative. The discipline is deciding what serves the so-what and what's simply interesting. A finding can be true, well-supported, and still not belong in this particular story if it doesn't move the argument forward — save it for an appendix, a follow-up analysis, or a different conversation entirely.

A practical filter: for each candidate finding, ask whether removing it would change the audience's understanding of the conclusion. If removing it changes nothing, it's decoration for this narrative, however much effort it took to produce. This is often the hardest discipline in storytelling — cutting analysis you're proud of because it doesn't serve this particular argument.

Anticipating Objections

A narrative that only presents supporting evidence invites the first skeptical question to derail the whole presentation. Think through the two or three most likely objections in advance — an alternative explanation for the trend, a data limitation someone will ask about, a segment that behaves differently — and address them proactively, briefly, before they're raised. This doesn't mean pre-emptively litigating every possible counterargument; it means handling the ones a reasonably informed skeptic in the room would actually raise.

Addressing an objection before it's asked signals that you've stress-tested the finding yourself, which builds more trust than a presentation that appears to have only considered evidence favoring its own conclusion.

Practical Review Checklist

Before moving on, confirm that you can:

• Write a one-sentence "so-what" for a piece of analysis you've done recently
• Order findings so each section answers "what does this lead to?"
• Apply the "would removing this change the conclusion" filter to decide what to cut
• Identify the two or three most likely objections to a specific finding
• Explain the difference between a narrative and a report

Conclusion

A strong data narrative isn't the analysis with the most findings — it's the one with the clearest throughline from question to evidence to recommendation, with everything unnecessary cut along the way.