.jpg)
Smart Analyses: Impactful Numbers
- Nov 11
- 10 min read
Drive Greater Impact & Inspire Action from Your Analyses (in 5 steps)
The thought and effort in the production of a meaningful analysis is an investment in time, effort and your thought. Think about the saying, "Genius is 1% inspiration and 99% perspiration" - nothing could be more true.
Imagine going through the pains to ensure your data's clean, the methodology's sound, and your insights could genuinely shift how your company approaches the next quarter. You are confident that what you've prepared will be embraced.
Then your stomach sinks as this scenario plays out: executive eyes glazing over by slide three, sidebar conversations starting, someone checking their phone. Twenty minutes later, you leave with "Let's circle back on this"—code for "we're not doing anything."
Here's the thing: your analysis wasn't the problem. Your presentation was.
I've seen brilliant analysts lose the room countless times, not because their work was weak, but because they buried the recommendation under fifteen slides of setup.
Meanwhile, I've seen mediocre analysis win budget and headcount because someone knew how to frame it for decision-makers. The difference isn't talent—it's structure.
This article shares my own strategies to turn any analysis into something executives embrace. We're talking about practical tactics you can use this week: writing for the person who signs off, cutting ruthlessly to what matters right now, leading with your answer instead of building to it, designing visuals that do the thinking for your reader, and closing with clear ownership so nothing falls through the cracks after the meeting ends.
Focus: Know Your Decision-Maker
Before you touch a single slide, answer this question: who has to say yes, and what do they care about?
I mean their actual job, not the org chart. A Chief Marketing Officer (CMO) managing return on ad spend (ROAS) in a specific demographic thinks differently than a Chief Financial Officer (CFO) worried about quarterly cash flow. Your Regional Vice President (VP) protecting their team's headcount has different pressure points than the Chief Product Officer trying to hit a launch date.
Let me give you an example. A streaming service needs to improve advertising performance for women aged 25 to 44. The analyst could start with an audience segmentation , walked through six testing methodologies, and eventually arrived at a recommendation. Instead, they opened with one sentence: "Shift 15% of budget to ad-supported video on demand (AVOD) to hit plus-10% ROAS in Q4."
That's it. Decision, owner, metric, timeline—all in 15 words.
The CMO knew immediately whether to keep listening. She could see her success metric (ROAS), her audience (women 25-44), her timeline (Q4), and the ask (shift 15%). Everything that followed just supported that opening call.
Here's how you build that opening sentence:
Name the decision-maker by role, not just department, know your target
Use their vocabulary—if they say "cost per acquisition," don't say "customer acquisition cost"
Tie to a metric they own and get measured on
Include the timeline that matches their planning cycle
Write it before you build anything else, then test every slide against it
You know what kills most presentations? Trying to be comprehensive. You think you're being thorough; they think you don't understand what matters. Your job isn't to show how smart or accomplished you are—it's to change what happens on Monday.
Make Your Analyses Relevant
I'm going to say something that might sting a little: most of your analysis doesn't belong in the deck.
That beautiful cohort breakdown? The sensitivity analyses that took two days? The competitive benchmark you pulled from three industry reports? If it doesn't alter the decision happening in this room, right now, it goes in the appendix or a follow-up email.
This isn't about dumbing things down. It's about respecting attention—the scarcest resource in any organization. Save the methodology and statistical calculations for the appendix, it distracts from the focus of the message. And by the way, the expectation is that you have done your due diligence - credibility is tied to your sourcing.
Think about it this scenario. A retail team is planning a holiday spend and originally built a 20-chart deck covering audience demographics, channel mix, creative performance, attribution methodology, seasonal adjustments, and competitive context. All solid work. But when they asked themselves "What actually changes the budget allocation this week?" only three charts mattered: incremental reach by channel, cost per incremental purchase, and the recommended spend shift.
They cut 17 charts. The meeting went from 45 minutes to 15. The budget got approved that afternoon.
Here's a framework that helps: Must/Should/Could.
Must = directly influences the decision and can't be inferred from anything else
Should = adds important context but the decision could happen without it
Could = interesting but doesn't change the outcome
Be ruthless with "Should" and "Could." Move them to backup slides or a separate document. Your main narrative should contain only "Must" items, and honestly? Most decks have three to five of those, maximum.
One more thing. Sometimes you discover something fascinating during analysis—a surprising correlation, an unexpected segment behavior, a methodological insight. That's exciting for you; it's a distraction for them. If it doesn't connect to the decision at hand, save it for a different conversation. There's always another meeting. Or, talk to the key stakeholder about what you discovered in a brief conversation prior to your meeting - think of it as a temperature check for their interest.
Lead with the Answer (Yes, Immediately)
Journalists learned this a century ago: put the most important information first. They call it the inverted pyramid—you lead with the conclusion, then provide evidence, then add context, with the least essential details at the bottom. Why? Because readers might stop at any point, and you want them to get the core message even if they only read two sentences.
Business presentations should work the same way, but somehow we got trained to do the opposite. We set context, explain methodology, show twelve charts, and finally—if people are still paying attention—reveal what we think they should do.
That's backwards.
Your opening slide should be the recommendation. Not the setup, not the agenda, not the executive summary—the actual call to action with enough specificity that someone could act on it immediately.
Imagine how this could play out. A fintech company proposes changes to customer onboarding. The analyst titled their presentation "Analysis of KYC (Know Your Customer) Flow Optimization." Generic, vague, forgettable.
Here's what they should have led with: "Shorter KYC flow raises funded accounts by 9%."
See the difference? The second version is a conclusion, not a topic. It tells you the outcome, not the subject. When you read it, you immediately know whether you care—and more importantly, whether you agree.
Each section should work the same way. Not "Q3 Performance Review" but "Q3 exceeded target by 12% despite pricing headwinds." Not "Regional Analysis" but "Mid-Atlantic delivers lowest cost per visit: protect that spend."
Here's your structure:
Answer → What should we do?
Insight → What did we learn that drives that answer?
Implication → Why does this matter to the business?
Recommendation → Specific actions with owners and dates
Next Step → How do we track whether it's working?
Make each section title something a leader could repeat verbatim in the next meeting. If your title needs explanation, it's not clear enough.
One tactic that consistently works: send a one-page pre-read 24 hours before the meeting. State the decision, lay out your recommendation, include one supporting chart. This sounds risky—won't people dismiss it before you present? Actually, the opposite happens. They come prepared to discuss trade-offs instead of hearing your logic for the first time. Meetings get shorter. Decisions happen faster.
Make Your Visuals Do the Heavy Lifting
Let me tell you what happens when someone puts up a chart with four y-axes, 12 data series, and a legend that requires a decoder ring: everyone stops listening to you and starts trying to figure out what they're looking at. You've just turned decision-makers into puzzle-solvers.
Bad visuals don't just fail to communicate—they actively work against you by burning mental energy on decoding instead of deciding. And, at that point, your group's attention has left the room
Edward Tufte, who basically wrote the book on data visualization, had a principle he called "data-ink ratio"—maximize the information, minimize the decoration. Every line, color, and label should help answer the question. If it doesn't, delete it.
Here's what that looks like in practice. A regional operations leader was reviewing cost per acquisition across eight markets. The original chart tried to show everything: monthly trends, year-over-year comparisons, budget versus actual, all on one busy graph with eight overlapping lines in different colors.
Instead, they used small multiples—eight simple charts, one per market, all with the same scale and format. Each market got its own mini-chart showing the trend, with one annotation: "Mid-Atlantic cost per acquisition down 18%: maintain spend."
Suddenly, the pattern was obvious. You could see at a glance which markets were improving, which were declining, and where to hold versus cut. The decision practically made itself.
Match your chart type to your question:
Comparing options? Bar chart, sorted by the metric that matters
Showing change over time? Line chart with minimal gridlines
Revealing relationships? Scatter plot with a clear trend line
Providing context? Small multiples that enable visual comparison
Use color for one thing only: directing attention to what matters most. Not to make it pretty, not because you're bored with gray—only to highlight the data point that drives your recommendation.
Here's a rule I follow: every visual gets exactly one annotated takeaway, written directly on the chart, in plain language, tied to the decision. Not "AVOD performance" but "AVOD delivers 22% more incremental visits per dollar: shift budget here."
When your annotation tells someone what to conclude, you control the narrative. When you make them figure it out, you've lost the room.
Close with Ownership and Honesty
You've stated the recommendation. You've shown the evidence. You're almost done—and this is where most people fumble.
They end with "Happy to discuss" or "Let me know if you have questions." Which is fine, I guess, if your goal is a polite conversation that leads nowhere. But if you actually want something to change? You need to get specific about who, what, when, and how we'll know if it worked.
Here's the template: present two or three options with trade-offs, pick one, name the owner, set a date, define the key performance indicators (KPIs), and add a brief methods note so everyone understands what could change the conclusion.
Let's break that down. For the purposes of illustration, let's say a B2B marketing team is recommending a channel shift. Their close looked like this:
Recommendation: Move 15% of budget from linear TV extensions to AVOD starting this week
Owner: Media Lead, with creative team support
KPI: Cost per incremental site visit, measured weekly
Check-in: Two-week readout on November 18
Source: Attribution uses last-click with 7-day window; seasonal adjustment based on prior two years; if holiday traffic patterns shift significantly, we'll reassess the model
That last piece—the source—is where your credibility lives. Everyone knows data has limits. Models make assumptions. Seasonality might surprise you. When you acknowledge that explicitly, you're not showing weakness; you're showing rigor.
Include a simple methods table: data sources, time window, definitions, key assumptions. Five lines, maximum. This isn't about being comprehensive—it's about flagging what could invalidate the recommendation so leaders can judge the risk themselves.
And then actually assign ownership. Not "the team will handle it" but "Sarah owns implementation, checks in every two weeks, reports cost per visit and incremental reach." When someone's name is attached, things happen. When it's vague, it drifts.
End each major section the same way: "What to decide now" and a checkpoint date. This keeps momentum going beyond the meeting and turns agreement into execution.
Putting It All Together (Your Checklist for This Week)
Honestly, you could read this entire article, nod along, and change nothing about how you present. Or you could pick one deliverable—maybe that quarterly review due Friday, maybe the budget proposal you're drafting—and actually apply this structure.
Here's what that looks like in practice:
Before you build anything: Write a one-sentence brief. "Improve Q4 ROAS in women 25-44; success equals plus-10% ROAS in six weeks; decision owner equals CMO." Everything you create gets tested against that sentence.
As you develop the narrative: Keep only the core insight that changes the call. "Prime-time AVOD supplies most incremental visits with lower cost per incremental outcome than linear extensions" justifies a budget shift. The twelve other insights you discovered? Appendix.
When you design visuals: Use small multiples by region or segment, with one callout on where to act. Clean, comparable, decisive.
When you close: State the recommendation (shift 15% budget), assign the owner (Media Lead), define KPIs (cost per visit), schedule a readout (two weeks), and include a methods note on attribution and seasonality that makes uncertainty explicit.
You know what's interesting? This approach feels riskier at first. You're exposing your recommendation immediately, before you've built the case. You're cutting material that felt important. You're committing to specifics that could be wrong.
But that discomfort is actually the point. When you make strong, clear, accountable recommendations, you force better decisions. The weak stuff gets challenged and improved. The strong stuff moves forward quickly.
What Actually Changes on Monday?
I started with a story about watching analysis get ignored in a meeting. Let me tell you how this scenario played out for the next presentation.
The team cut their 22-slide deck to four slides. They opened with the recommendation in the first sentence. They used one annotated chart per key point. They ended with specific owners, dates, and KPIs. The meeting took 12 minutes. The CFO approved budget that afternoon. Implementation started the following Monday.
Nothing about the underlying analysis changed. The data was the same, the methodology identical. What changed was the packaging—or really, the respect for how decisions actually get made in organizations.
Here's your challenge for this week:
Take your current project—whatever you're working on right now—and rewrite the opening to include the decision, owner, and success metric in one sentence. Then look at every slide and ask "Does this change the call?" If not, move it or delete it. Pick your densest chart and convert it to small multiples with one clear annotation. Finally, draft your close with two options, a chosen recommendation, an owner, a date, and a five-line methods note.
Do that, and I guarantee your next meeting goes differently. Not because your analysis improved—because you finally presented it in a way that respects how busy, skeptical, impatient decision-makers actually absorb information and commit to action.
Every deliverable should answer one question: "What will change on Monday because of this?" If you can't answer that clearly, in one sentence, before you build your deck? You're not ready to present yet.
Now go write that winning presentation.
Ready to level up your data game? Let's make it happen! 🚀
💡 Need strategic insights for your next project? Let's collaborate as your analytics consultant.
🎤 Looking for a dynamic speaker who makes data come alive? Book me for your next event.
📈 Want to master the art of analysis yourself? Reach out to learn my proven strategies.
Your data has stories to tell – let's unlock them together!
