To end a data presentation, swap "any questions?" for a call to action that names three things: the decision you're recommending, the person who owns it, and the date it happens by. Action, owner, deadline, in one sentence. I've watched sharp analysts spend six weeks on a model, then close with a slide that says Thank You. Six weeks of work. Thirty seconds of ending. Nothing decided.
Why this matters (and what most executives get wrong)
The end is the moment of peak attention. People are leaning in. Most presenters spend it thanking everybody. Psychologists call the habit of judging an experience by its high point and its final moments the peak-end rule, and a presentation that fades out gets remembered by the fade.
Numbers make it worse. Your conclusion feels obvious after a month inside the data, so you assume the room arrives there too. They won't. They saw the chart for eleven seconds. Setup, conflict, resolution is the three-part cinematic structure for a data story, and the call to action is the resolution. The general craft is covered in how to end a presentation. This one is narrower: what do you ask for when the deck was numbers?
The 5-step close that turns numbers into a decision
I write the last ninety seconds before I build a single slide.
1. Write the decision your data already narrowed
Your numbers didn't only describe a situation, they ruled things out. Start there. What does this data make it hard to keep doing? Write it as one sentence beginning "I'm recommending that we." Can't finish it? No ending yet.
2. Name the owner, and name a person
"The operations team will handle implementation" is not an owner. It's fog. Somebody in that room, quite possibly the person in seat three, has to nod, and a request handed to a department instead of a person quietly dies in the meeting notes.
3. Put a date on it
End of quarter two. First week of June. Dates feel bureaucratic when you type them and sound like leadership when you say them, because a date separates a recommendation from a plan.
4. Shrink the ask until yes is easy
This is the one I'd argue hardest for. Don't ask for the whole program. Ask for the first move: approve the pilot at one site, fund eight weeks, let one team report back. Big asks get deferred. Small asks get decided.
5. Signal the close, then say it once
Give the room a runway. "As I begin to wrap up, here are the last pieces" tells people the ending is arriving. Then deliver the ask and stop talking, which is harder than it sounds.
"Based on what we covered today, I'm recommending we approve the reallocation, task operations with the new process by the end of quarter two, and set a checkpoint the first week of June."
How to end a data presentation when there is nothing to decide yet
Sometimes the honest answer is that the data isn't ready. Fine. That's no reason to trail off. The close changes shape: here is the number we're watching, the line that would change our minds, and when we look again. Name who watches it and put the next look on the calendar before the room breaks up. A meeting that schedules its own next decision has still produced an output.
The mistake most executives make
The pattern I see constantly is a recommendation that sits next to the data instead of coming out of it. Eighteen slides of methodology, then slide nineteen arrives with three bullets that could have been written before a single query ran. The room feels that seam. Your ask has to be the thing the numbers force. If your recommendation would survive a completely different data set, it isn't coming from your data.
Case study: the quarterly briefing that finally decided something
Anonymized but the pattern repeats often in the work I do... Picture a research analyst inside a state agency giving the same quarterly program briefing: enrollment by county, cost per participant, twenty-plus slides, one unvarying closing line. "So those are the numbers, happy to take questions." Nothing ever changed.
We rewrote the last ninety seconds, nothing else. She recommended shifting outreach hours into the two counties her data kept flagging, named the program manager sitting a few seats away, and asked for a decision inside two weeks. The room argued, narrowed it to one county, approved that. The argument was the win.
| How you close | What the room can do | What happens |
|---|---|---|
| "Thank you. Any questions?" | Ask something, or nothing | Nobody owns anything |
| "So those are the findings." | Nod politely | Topic returns next quarter |
| "I recommend this, she owns it, by Friday." | Approve, modify, push back | The meeting produces an output |
| "Nothing to decide yet. Here is our tripwire." | Accept it or set another | The next decision gets a date |
Frequently asked questions
How do I end a data presentation if the decision isn't mine to make?
You still make a recommendation. Say what the data points to, then ask the person who owns the decision to accept it, change it, or reject it. Name that person out loud. Numbers with no recommendation leave the room with direction from nobody.
What if my data doesn't support a clear recommendation yet?
Then ask for attention instead of approval. Name the number you're watching, the threshold that would change your recommendation, who is monitoring it, and when the group looks again. Put that review on the calendar before anyone stands up. It is still a call to action.
Should I still take questions at the end of a data presentation?
Yes, but not as your ending. Take questions before the close, then deliver the call to action last so the ask is the final thing anybody hears. End on questions and whatever the last person asked becomes your closing note. You handed your ending to a stranger.
How long should the closing call to action be?
Three sentences is usually enough. One names the decision, one names the owner, one names the date. A close running much past thirty seconds starts adding qualifiers, and qualifiers give the room permission to defer. Say it, stop talking, and let the silence work.
What to do next
Open your next deck and jump to the final slide. If it says Thank You or Questions, that's a stopping point, not an ending. Write three sentences: decision, owner, date. Then cut the ask in half.
The ending is the resolution of the whole story, which is why it has its own place inside Story-Driven Data™. Start with what data storytelling really is, read the general anatomy of the ask in presentation call to action, and if a high-stakes numbers presentation is on your calendar, get a quick quote.