Before you build a single slide, ask one question: what decision does this data need to support? Not what the data shows, not what is interesting in it, but what choice it is supposed to help someone make. That one question is the most useful of all the data presentation tips I teach, because it changes what you keep, what you cut, and how you stand in the room.
I watched a finance director open a quarterly review with forty-one slides. By slide nine the CEO had stopped looking up. She had the numbers cold. What she did not have was an answer to the only question her boss actually carried into the room: do we fund the second plant or not? The data was all there. The decision was nowhere.
Why this matters (and what most executives get wrong)
Most data presentations fail before anyone opens PowerPoint. They fail at the goal. People sit down to build and silently set the target as "show what the data says." That feels responsible. It is also the trap.
When showing is the goal, every chart earns a place. The regional breakdown, the year-over-year, the cohort view, the appendix you spent a weekend on. Nothing gets cut, because everything is technically true. So you walk in with thirty slides and walk out with no decision.
Here is what I tell clients. Your audience did not block ninety minutes to admire your spreadsheet. They came to decide something, or to feel confident that someone has. The numbers are evidence. Evidence is in service of a verdict. When you skip naming the verdict, you hand the room a pile of evidence and ask them to assemble the case themselves, in real time, while you talk. They will not. They will get quiet, then defer.
This is the part nobody wants to hear: a flawless chart attached to no decision is still a failed slide. Accuracy is not the bar. Usefulness is.
Your audience did not block ninety minutes to admire your spreadsheet. They came to decide something.
The four-step process to name the decision before you build
This is the front end of data storytelling, and it is part of the broader Story-Driven Data™ approach I use with executives. You do not need the whole system to start. You need four steps, in order, before you touch a slide.
Step 1: Write the decision in one sentence
Before any chart, finish this sentence on paper: "After this presentation, the room should decide whether to ____." Fund the plant. Kill the SKU. Move the launch to Q3. If you cannot finish the sentence, you are not ready to build. You are ready to go ask whoever called the meeting what they actually need to walk out knowing.
Step 2: Find the one number that moves that decision
Every decision turns on a single number more than any other. Call it the hero number. It is the one figure that, if it moved, would change the answer. For the plant, it might be projected capacity utilization, not revenue, not headcount, not the twelve other metrics on the dashboard. Find that number and build the whole presentation to deliver it cleanly.
Step 3: Write the "which means" sentence for every chart
For any chart that survives, write one sentence underneath: "which means." Sales dipped 8% in the Northeast, which means we are losing the accounts we assumed were locked. The chart is the what. The "which means" sentence is the so-what, and the so-what is what your audience can actually use. A chart with no "which means" line is decoration. Cut it or finish it.
Step 4: Build a case, not a report
Now structure it as setup, conflict, resolution. Setup: here is the decision and the number that drives it. Conflict: here is the tension, the trend going the wrong way, the two options that both look reasonable. Resolution: here is what the data says to do, and here is my recommendation. A report lists what happened. A case argues toward a decision. You want the case.
Take your next deck and write the decision sentence at the top of slide one, in plain words, before you edit anything else. Then delete every slide that does not help the room make that one choice. Most people cut a third of their slides in ten minutes and feel relieved, not anxious.
The mistake most executives make
The mistake is confusing the role of reporter with the role of advisor. A reporter shows what happened and stops. Accurate, complete, neutral, and forgettable. An advisor says what it means and what to do about it. They put a recommendation on the table and let the data back it up.
Executives slide into reporter mode because it feels safer. If you only present the facts, no one can argue with your judgment, because you offered none. But that safety is exactly why you get talked over and second-guessed. The room cannot tell whether you understand the decision or just compiled the inputs.
Naming the decision up front does something else worth noting. It keeps people in their seats. When the room knows a decision is coming, attention holds, because everyone is waiting to see whether they agree. When no decision is coming, attention leaks by slide nine. Every executive in your audience is quietly triaging their inbox against the value of your next chart. Give them a verdict to anticipate and you win the triage. This is also why so many status meetings feel like a waste. The research backs the instinct, and Harvard Business Review has written about the real cost of bloated meetings in Stop the Meeting Madness. A presentation with no decision attached is just a meeting that bills by the slide.
Case study: the CFO who opened with the bad news on purpose
Here is a real one. The average chief financial officer holds the job for around eighteen months, because when the numbers slip, the CFO is the easy person to blame. So the pressure is highest exactly when the data is worst. I worked with a CFO at a healthcare company, which I will keep anonymous, who had to walk leadership through a quarter where earnings were down and the story was not pretty.
The usual advice is to bookend the bad news: good news first, the miss in the middle, good news on the way out. We made a different decision, and we built the entire deck around it. He opened with the hard truth. Things were off, earnings were down, last quarter was a problem. Then he paused, looked around the room, and asked, "Are you ready for the good news?" The whole room smiled. Now his real presentation could begin.
Putting the elephant in the room first showed leadership, confidence, and courage. From there he laid out the factors behind the miss and a clear plan for the next few quarters. The team left energized and, more important, certain they were in good hands. One decision, confront it first, reorganized the entire presentation.
Reporter versus advisor: the same data, two outcomes
| Dimension | Reporter | Advisor |
|---|---|---|
| Goal of the deck | Show what the data says | Support a specific decision |
| Slide one | Agenda or background | The decision in one sentence |
| What gets included | Everything that is true | Only what moves the decision |
| Each chart | Stands alone as a fact | Carries a which-means line |
| Structure | A report: what happened | A case: setup, conflict, resolution |
| How the room responds | Quiet, then let's circle back | Engaged, then a decision |
Look at the table for a second. Nothing in the advisor column requires better data than the reporter column. Same numbers. Same charts, mostly. The difference is a question asked before the building started, and the discipline to let the answer cut everything that does not serve it.
What to do next
Start with your next deck, not your next quarter. Open the file, go to slide one, and write the decision sentence at the top before you change anything else. Then run the four steps: name the decision, find the hero number, write a "which means" line under every chart, and rebuild it as a case instead of a report. You will cut slides, and the cutting will feel good, because for the first time you have a rule for what stays.
If you want help turning your team's data into decisions the room actually makes, get a quick quote and we will talk about what your presentations need. To go deeper on the method behind this, read what is data storytelling and see how the pieces fit together.
One question, asked before you build. It is the cheapest upgrade you will ever make to a data presentation, and it costs you nothing but the discipline to answer it honestly.