Warm Up
Discuss these questions with your partner before you start the lesson.
Question 1
Think of the last time you presented data to a senior audience. Did they engage with the analysis — or did they skip to 'what does this mean for us?'
Question 2
Have you ever seen someone confuse correlation with causation in a business meeting — concluding that one thing caused another just because they happened at the same time?
Question 3
When you look at a chart or graph in a presentation, how long does it take you to understand it? What makes some visualisations immediately clear and others confusing?
Vocabulary
8 key words for this lesson
So what
The question that turns a data point into an insight. 'So what?' asks: why does this number matter? What does it mean for the business? Every data slide needs a clear answer to this question.
Riley looked at Marcus's slide: Conversion rate is up 3%. Riley said: So what? What does that mean in revenue terms? What caused it? What should we do next? If your audience is asking these questions, the story is not done yet.
Insight pyramid
A storytelling framework that starts with the recommendation at the top, supported by key insights, each supported by data. It is designed for busy audiences who want the answer first, not the analysis.
Riley said: Busy executives do not want to watch you build to the conclusion. Use the insight pyramid — give them the recommendation first, then the three insights that support it, then the data underneath. Answer first, evidence second.
Data narrative
The story built around data — connecting numbers, insights, and implications into a logical sequence that leads to a clear conclusion. A data narrative is not a data dump — it is a structured argument.
Marcus said: I have 40 slides of data. Riley said: That is not a data narrative — that is a report. A narrative has a beginning, a middle, and an end. It has tension and resolution. What is the one thing you want the audience to believe when you finish?
Visualisation
The way data is displayed — as a chart, graph, table, or diagram. Good visualisation makes a pattern obvious in seconds. Bad visualisation makes people stare at a slide and still not understand.
Riley said: A bar chart comparing five numbers is clear. A table with 40 rows is not — that is a data dump, not a visualisation. The test of a good visualisation is whether someone can understand the point in under five seconds without you explaining it.
Inference
A conclusion drawn from data — going beyond what the numbers say directly to explain what they mean or predict what comes next. In data storytelling, inference is how you turn observation into insight.
Riley said: The data shows churn went up in Q3 and NPS dropped in Q2. The inference is that the satisfaction problem in Q2 caused the churn in Q3 — they are connected. That is the insight. Now we know where to focus.
Headline metric
The single most important number that summarises the state of a business or campaign. A good headline metric is simple, meaningful, and immediately understood by a non-expert audience.
Riley said: Your board does not want 20 metrics. They want one headline metric — the number that tells them if the business is healthy or not. For us, that is monthly active users. Everything else is context.
Correlation vs causation
The difference between two things happening at the same time (correlation) and one thing causing the other (causation). Confusing them leads to wrong conclusions and bad decisions.
Marcus said: Ice cream sales and drowning rates both go up in summer — does ice cream cause drowning? Riley said: No — that is correlation, not causation. Both go up because of hot weather. When you present data, always ask: are we observing a pattern or explaining a cause?
Benchmarking
Comparing your results to an external standard — such as an industry average, a competitor, or a previous period. Without benchmarking, a number has no context and no meaning.
Riley said: Our open rate is 22%. Is that good? Without benchmarking, I cannot tell you. Against an industry average of 18%, it is great. Against our top competitor who is at 35%, it is a problem. Context changes everything.
Phrases
6 phrases and their meanings
The story the data is telling us is...
A phrase that frames data as a narrative rather than a collection of numbers. It signals that you have done the analytical work and are now presenting a clear conclusion, not a data dump.
Riley opened her presentation: I am not going to walk you through every number. The story the data is telling us is this: we are winning on acquisition but losing on retention — and retention is where the revenue lives. That is the problem we need to solve.
What this means for the business is...
A connecting phrase that turns a data observation into a business implication. It is the 'so what' made explicit — bridging the gap between the number and the decision.
Riley said: NPS dropped 8 points in Q2. What this means for the business is that churn will increase in Q3 unless we act now. That is the implication. The data is the evidence. The action is what we decide today.
Let me give you the headline first
A signal that you are structuring your presentation using the insight pyramid — answer first, evidence second. It shows you understand that your audience wants the conclusion before the analysis.
Riley said: Let me give you the headline first — our customer acquisition cost has increased by 40% and our payback period is now 18 months. That is the problem. Now let me show you the three reasons that explains why and what we recommend.
The inference we draw from this is...
A phrase that signals you are moving from data observation to analytical conclusion. It is what separates a data presenter from a strategic thinker.
Riley said: Engagement dropped 30% in the week we changed the homepage. The inference we draw from this is that the new design is hurting discovery. That is a hypothesis, not a certainty — but it is where we look first.
We need to be careful about correlation versus causation here
A phrase that signals analytical discipline — warning the audience not to assume that because two things happened at the same time, one caused the other.
Marcus said: Sales went up 20% in the same month we launched the campaign. Riley said: We need to be careful about correlation versus causation here. Did the campaign drive the sales, or was there a seasonal factor? We need the controlled data before we claim credit.
To put that in context...
A phrase used to add benchmarking or comparative data that gives a number meaning. Without context, a number is just a number — this phrase is how you make it a story.
Riley said: Our churn rate is 5% per month. To put that in context — the industry average is 2%. We are losing customers at more than twice the rate of our competitors. That context is what turns a number into an urgent problem.
Videos
Watch these terms used in context
Dialogue
Read the dialogue and hover the blue words for definitions
Marcus
RileyRiley, I have a 45-minute slot in front of the board next week to present the marketing performance data. I have 40 slides. Where do I start?
Delete 35 of them. A board does not want 40 slides of data — they want a data narrative. What is the one thing you want them to believe when you finish?
I want them to approve more budget for retention activities. We are losing customers and the data shows it clearly.
Good — that is your headline metric. Lead with it. Let me give you the headline first: our monthly churn is 5% — more than double the industry average. Everything else supports that.
Should I use the insight pyramid?
Yes. Recommendation first. Then the three insights that support it. Then the data underneath each insight. Board members are busy — they want the answer before the evidence.
One of my insights is that NPS dropped in Q2 and churn went up in Q3. I think the satisfaction problem caused the churn.
Good inference. But be careful — we need to be careful about correlation vs causation here. Was there another factor in Q3? Acknowledge the hypothesis, but do not overclaim.
Good point. I also want to show the churn rate in context against competitors.
That is benchmarking. Use the phrase: 'to put that in context — the industry average is 2%, we are at 5%'. Context is what makes a number urgent instead of just interesting.
And for my visualisation — I have been using big tables. Is that okay?
No. Tables are for reading, not presenting. Use one number or one chart per slide. The story the data is telling us should be visible in under five seconds. If someone needs to read a table to understand it, the slide has failed.
Okay — one chart per slide, headline first, insight pyramid structure. What else?
End with the so what. What this means for the business is: we need an extra £200k in retention investment now, or we will spend £800k replacing those customers next year. That is your close.
Exercises
Complete all three exercises to see your final score
Complete the Sentence
Choose the correct answer to complete each sentence.
Choose the correct word or phrase to complete each sentence. Only one answer is correct.
1.Before I walk through the numbers, let me give you the headline .
2.Ice cream sales and drowning both go up in summer — that is , not causation.
3.Our churn rate is 5%. To put that in — the industry average is 2%.
4.The we draw from this data is that the NPS drop in Q2 caused the churn spike in Q3.
5.The board wants one metric — the number that tells them immediately if the business is healthy.
6.A table with 40 rows is not a visualisation — it is a data .
Matching
Click a word, then click its correct definition.
Click a word on the left, then its definition on the right.
Words
Definitions
Fill in the Blank
Drag the correct words from the word bank to complete each sentence.
Sentence 1
Recommendation first, evidence second — that is the — — — pyramid approach.
Sentence 2
To put that in — — — — the industry average is 2% and we are at 5%, more than double.
Sentence 3
The story the — — — is telling us is that we are winning on acquisition and losing on retention.
Sentence 4
Be careful about correlation versus — — — — two things happening at the same time does not mean one caused the other.
Sentence 5
What this means for the — — — is that we need to act now or the churn problem will cost us three times more to fix later.
Sentence 6
The — — — we draw from this is that the NPS drop in Q2 caused the churn spike in Q3.
Tip: Think about the meaning of each word and what makes sense in the sentence.
Multiple Choice
Choose the best answer for each question about the dialogue.
1.What does Riley say is the main problem with 40 slides of data?
2.What does Riley recommend as the headline metric for Marcus's presentation?
3.Why does Riley say to be careful about correlation versus causation?
4.What does Riley say is the test of a good visualisation?
Group Activities
Role-play scenarios and discussion questions for group classes
🎯 Choose the Best Response
Read what the senior stakeholder says. Choose the best response from the three options.
Customer says
“You've shown me a lot of numbers. What is the point you are trying to make?”
Salesperson responds…
Customer says
“Sales went up 20% the same month we launched the new campaign. Can we say the campaign caused it?”
Salesperson responds…
Customer says
“This slide has too many charts. Which number should I focus on?”
Salesperson responds…
🔍 Spot the Mistake
Read the conversation. Three lines have a mistake. Can you find them?
Marcus
I'm going to present the data in a logical order — starting with the background, building to the insights, and ending with the recommendation.
Riley
For a board audience, that is the wrong structure. Use the insight pyramid — start with the recommendation, then show the evidence.
Marcus
Understood. And I'll benchmark our results — to put that in perspective, our churn is 5% versus an industry average of 2%.
Riley
The fixed phrase is 'to put that in context', not 'in perspective'. Context is the standard for benchmarking language.
Marcus
Noted. And I'll be honest about the limits — sales and campaign launch happened at the same time, so I'll say there is causation between the two.
Riley
No — you can say there is correlation, not causation. We need controlled data before we claim one caused the other.
✏️ Finish the Salesperson's Line
The data storyteller starts a sentence. Work with your partner to finish it naturally.
Customer
“I have all this data but I don't know how to make it into a story for the board.”
Salesperson
“Start with the one thing you want them to believe and act on when you finish — that is the headline. Then build the insight pyramid: recommendation at the top, three supporting insights in the middle, and the data underneath. The story the data is telling us is... …”
Customer
“Our engagement went up the same month we changed the pricing. Can we say the pricing drove it?”
Salesperson
“We need to be careful about correlation versus causation here. Both happened in the same month, but that does not mean one caused the other. Before we make that claim, we need to check for other variables — seasonal trends, a product update, a competitor move. The inference we draw from this is... …”
Customer
“The slide is confusing. There are too many numbers on it.”
Salesperson
“You are right — that is my fault. One number, one chart, one insight per slide. Let me give you the headline first: …”