Lesson 12 of 16 · Level 3: Advanced
Data Storytelling: Presenting Your Findings
- ⏱ 12 min read
- ❓ 4 review questions
- ✍️ Exercise with solution
- 📥 Downloadable template
In this lesson
- Why findings without a story rarely lead to decisions, and what a "big idea" is.
- A simple three-part structure for any data story: what, so what, now what.
- How to adapt the same findings for executives, managers and fellow analysts.
- How to storyboard a five-slide presentation with headline titles, using the course dataset.
- How to present limitations and uncertainty honestly without weakening your message.
Every number in the story below comes from the practice data. Use the Summary sheet to build the charts for your own version of the presentation.
- sales-practice-data.xlsx · sheets Sales Clean and Summary
Data storytelling is the skill of turning analysis into a short, clear message that a specific audience understands, believes and acts on. It combines three things: the right data, a few well-chosen visuals and a narrative that explains why it matters. You can do flawless analysis and still fail if your manager walks out of the meeting unsure what you found. In this lesson you'll build a complete story from the course dataset, from the one-sentence big idea to a five-slide deck.
Start with the big idea
Before you open PowerPoint or Google Slides, write the single sentence you want your audience to remember. This "big idea" should state your point of view and what's at stake. For our retailer's year-end review:
Our 2025 revenue depends on a few large wholesale orders, so we should build a program to win and keep wholesale accounts, while continuing to grow steadier consumer sales.
If you can't write the sentence, you're not ready to present: you have findings, not a message. Everything that follows (charts, numbers, slides) exists to support that sentence. Anything that doesn't support it goes to the appendix.
The structure: what, so what, now what
Most effective data stories follow the same arc, whether it's a two-line email or a 20-minute presentation:
- What?The findingState the fact with a number. "Wholesale was 18.5% of orders but 68.5% of revenue in 2025."
- So what?Why it mattersExplain the implication. "Five orders produced nearly a quarter of revenue. Losing one or two large buyers would cut our revenue sharply."
- Now what?The recommendationSay what to do, who should do it and by when. "Create a wholesale account program in Q1 and track the pipeline monthly."
Lead with the answer. Executives often call this BLUF (bottom line up front): give the conclusion first, then the evidence. Analysts love to tell the story in the order they discovered it ("first I cleaned the data, then I built a pivot…"), but your audience doesn't need your journey; they need the destination and enough evidence to trust it.
1. Which opening line follows the "bottom line up front" principle?
BLUF puts the conclusion and recommendation first. The method belongs later (or in an appendix), and "many interesting patterns" gives the audience nothing to hold onto.
Know your audience
The same findings need different packaging depending on who's listening:
- Detail wanted
- ●●●●
- Time available
- ●●●●
- Give them
- The decision, the impact in dollars and the risk. 3–5 slides or a half-page memo.
- Detail wanted
- ●●●●
- Time available
- ●●●●
- Give them
- Their own slice (region, channel), what changed and what they can do about it.
- Detail wanted
- ●●●●
- Time available
- ●●●●
- Give them
- Methods, definitions, cleaning log, queries and limitations so they can check your work.
Show one idea per chart
The central finding of our story is a contrast between orders and revenue. Two simple 100% bars make it impossible to miss:
Share of orders (200 orders)
Share of revenue ($89,735)
Headline: "Wholesale is fewer than 1 in 5 orders but more than two-thirds of revenue." Online is 17.7% of revenue and retail 13.8%.
Notice how the design does the storytelling: wholesale is the only colored segment, the consumer channels are gray because they're context, and the headline says exactly what to look at. Your audience shouldn't have to discover the insight; you hand it to them.
Storyboard: the five-slide year-end review
Sketch every slide before building any of them. Each one gets a headline that states its point (a full sentence, not a topic) and a single visual:
| # | Headline | Visual | Key numbers |
|---|---|---|---|
| 1 | 2025 brought in $89,735 from 200 orders at a 47.4% gross margin | Four KPI cards | $89,735 · 200 · $448.68 · 47.4% |
| 2 | Wholesale is 18.5% of orders but 68.5% of revenue | Two 100% bars | 37 orders, $61,479 |
| 3 | Five orders produced almost a quarter of the year's revenue | Top-5 order bar chart | $21,454 = 23.9% |
| 4 | Consumer sales grew 42% in the second half and are far steadier | Line chart, consumer vs. total | H1 $11,679 → H2 $16,577 |
| 5 | Recommendation: launch a wholesale account program in Q1 2026 | Three actions with owners and dates | Success metric: wholesale orders per month |
Read only the headlines from top to bottom: they tell the whole story by themselves. That's the test of a good storyboard. Details such as the cleaning log, regional tables and SQL queries go in an appendix for anyone who asks.
2. Which slide title is most effective?
A headline title states the takeaway with a number, so the slide makes its point even if nobody reads the chart. The others are labels that leave the audience to figure out the message.
An example scenario
Jordan's first draft had 14 slides in the order the analysis was done: data sources, cleaning, a table per region, a chart per month. The VP's feedback: "What do you want me to do?" Jordan rebuilt it as five slides around one comparison: who places the orders vs. who brings in the money.
Same data, different story: the second version ended with a clear decision to discuss, in 10 minutes instead of 40.
Be honest about uncertainty
Storytelling is persuasion, which makes honesty non-negotiable. Two habits keep you credible:
- State the limitations in one line. "This is one year of data; we can't yet tell whether the December wholesale orders repeat. There's no customer ID, so we can't measure repeat buyers." Limitations don't weaken a story; hiding them does, when someone finds them later.
- Don't cherry-pick comparisons. "Revenue grew 506% from January to December" is technically true ($3,025 to $18,319), but it compares the year's weakest month with a month inflated by six bulk orders. An honest comparison uses consistent periods (H1 vs. H2, or this year vs. last year) and explains outliers.
"Wholesale is 68.5117% of revenue" suggests more certainty than the data supports. Round to what matters for the decision: 68.5%, or even "about two-thirds" when you're speaking.
3. A colleague wants to open with "Revenue grew 506% from January to December!" What's the problem?
The calculation is correct ($18,319 ÷ $3,025 ≈ 6.06, a 506% increase), but the chosen endpoints make the trend look far bigger than it is. Compare consistent periods and explain the December wholesale spike.
Delivering the story
- Say the headline before showing the chart, then point to the part of the chart that proves it.
- Translate numbers into business terms: "two-thirds of revenue from fewer than one in five orders" sticks better than two percentages.
- Anticipate questions: prepare a backup slide for the obvious ones ("Which customers are these?" "Is this normal for December?").
- End with the ask: the decision you need, and the next step you'll take.
Storytelling mistakes
Presenting the analysis in the order you did it.
Start with the conclusion, then the evidence.
Topic titles like "Regional analysis."
Headline titles that state the finding with a number.
Every chart you made, in case someone asks.
One chart per key point; the rest in an appendix.
Ending with "Any questions?"
End with the recommendation and the decision you need.
4. Your audience is the West regional manager. Which detail is most relevant to include?
Managers need their own slice and actionable detail. Methods are for analysts (put them in an appendix), and company-wide totals alone don't tell the manager what to do in their region.
Practice exercise
Write a three-sentence story (what, so what, now what) for the West regional manager, using these facts from the practice data: West revenue $36,626 (40.8% of the company); 14 wholesale orders brought in $28,131; the other 50 West orders brought in $8,495; California $15,467 and Arizona $9,776 lead the region; Q1 West revenue was only $593.
See solution
What: "The West was our largest region in 2025 with $36,626 (40.8% of revenue), but 76.8% of it came from just 14 wholesale orders."
So what: "That makes the region's results lumpy (only $593 in Q1) and dependent on a few buyers, mainly in California and Arizona, which together brought in $25,243."
Now what: "I recommend assigning an account manager to the top wholesale buyers in California and Arizona by the end of Q1, and setting a monthly target for consumer orders to smooth out the slow quarters."
Check the numbers: $28,131 ÷ $36,626 = 76.8%; $28,131 + $8,495 = $36,626; $15,467 + $9,776 = $25,243.
Summary
- Write your big idea in one sentence before building any slide.
- Structure every story as what, so what, now what, with the bottom line up front.
- Adapt detail and format to the audience: executives decide, managers act, analysts verify.
- Storyboard with headline titles; one idea and one visual per slide.
- State limitations, avoid cherry-picked comparisons and round sensibly.
Next lesson: Correlation, Trends and A/B Testing. You'll learn to measure relationships, spot misleading correlations and read the results of an A/B test.
Create your account in 30 seconds, mark your lessons and download your certificate at the end.