His numbers said 2.3%. Mine said 4.1%. Same data set. Same quarter. Twelve executives staring at both of us.
Quick Answer: When someone contradicts your data in a presentation, the instinct is to defend your numbers immediately. Don’t. The moment you argue about data in front of decision-makers, you’ve turned a presentation into a debate — and both sides lose credibility. Instead, use the Parallel Truth framework: acknowledge the discrepancy, identify the methodological difference that explains it, and redirect to the decision. Most data contradictions aren’t about someone being wrong — they’re about different assumptions producing different numbers from the same underlying data.
In this article:
- Why data contradictions happen (and why neither person is usually wrong)
- The Parallel Truth Framework: both numbers can be right
- The 3-step recovery: Pause → Identify → Redirect
- The 4 worst responses (and why smart people default to them)
- How to prevent data contradictions before they happen
- When you ARE wrong: the 15-second correction that builds trust
- Frequently asked questions
At Royal Bank of Scotland, I was presenting a quarterly risk review to the credit committee. Slide 6: default probability across the commercial portfolio. My analysis showed 4.1% — a number I’d built from our internal risk models over three weeks of work.
The head of credit risk interrupted. “That doesn’t match our figures. We’re showing 2.3%.”
Twelve executives looked at me. Then at him. Then back at me.
My instinct was to defend my methodology. I’d spent three weeks on this. I knew my numbers were right. But the moment I said “my numbers are correct,” I’d be calling the head of credit risk wrong — in front of his peers and his boss.
Instead, I said: “That’s useful. Can I ask — is your 2.3% based on the Basel II standardised approach, or the internal ratings model?”
He paused. “Standardised.”
“That explains the difference. My 4.1% uses the internal ratings-based model, which captures the concentration risk in the commercial book. Both numbers are accurate — they’re measuring the same portfolio through different lenses.”
The committee chair nodded. “Which lens should we be using for this decision?”
That question — not the data contradiction — was the moment that mattered. And I had the answer ready.
⚡ Presenting data to a room that might challenge it? 3 pre-meeting checks:
- ☐ Know your methodology in one sentence: “This uses [model/system/period/assumption]”
- ☐ Identify the 2-3 people most likely to have different numbers — and find out what data source they use
- ☐ Prepare the redirect: “Both numbers inform the decision. Here’s which one matters for THIS choice…”
🚨 Presenting data that someone in the room might challenge? Quick check: Can you explain your methodology in one sentence? Do you know who has different numbers and why? If not, you’re walking into a credibility ambush. → Need the complete data defence framework? Get the Executive Q&A Handling System → £39
Why Data Contradictions Happen (And Why Neither Person Is Usually Wrong)
Most people assume a data contradiction means someone made a mistake. That’s rarely true in executive settings. What’s usually happening is one of three things:
Different time periods. Your Q3 data runs July-September. Theirs runs August-October. Same metric, overlapping but different windows. Both are “Q3” depending on the reporting system. Both are accurate. Neither is wrong.
Different methodologies. You used an internal model. They used an industry benchmark. You measured gross. They measured net. You included contingent liabilities. They excluded them. The underlying data is identical — the analytical lens is different.
Different scope. Your analysis covers the entire portfolio. Theirs covers the top 50 accounts. You’re measuring company-wide. They’re measuring their division. Your “revenue growth” includes acquisitions. Theirs is organic only.
In 24 years of banking, I can count on one hand the number of times a data contradiction in an executive meeting was caused by an actual error. The other 95% of the time, it was a methodological or scope difference that produced different numbers from the same reality.
Understanding this changes everything. Because if both numbers can be right, the correct response isn’t to defend yours — it’s to identify why they’re different and redirect to which one matters for the decision at hand.
What should you do when someone says your data is wrong in a meeting?
Don’t defend immediately. The first 10 seconds determine whether this becomes a credibility crisis or a credibility win. Pause. Acknowledge the discrepancy calmly: “That’s a different figure — can I ask what data source that’s from?” This buys you 30 seconds to think, identifies the methodological difference, and positions you as collaborative rather than defensive. In most cases, both numbers are right — they just measure different things. The person who identifies WHY they’re different earns more credibility than either set of numbers.
The Parallel Truth Framework: Both Numbers Can Be Right
The Parallel Truth Framework is the foundational principle for handling data contradictions. It’s based on a simple reality: in complex organisations, the same data can produce legitimately different numbers depending on the methodology, time period, scope, and assumptions applied to it.
The principle: Instead of arguing about which number is correct, identify the assumption difference that explains both. Then redirect to which measurement framework matters for the specific decision the room is making.
The script: “Both figures are measuring [the same thing]. The difference is [specific assumption/methodology/scope]. For the decision we’re making today — [specific decision] — the figure that matters is [X] because [specific reason].”
Why it works: It does four things simultaneously. It validates the other person’s numbers (no conflict). It demonstrates deep understanding of the data (credibility). It identifies the methodological nuance (expertise). And it redirects to the decision (leadership). The person who explains the discrepancy looks like the smartest person in the room — regardless of whose number is “right.”
This framework is part of a broader system for handling difficult questions in presentations — but it requires specific preparation for data challenges that general Q&A frameworks don’t cover.
⚡ Your 15-Second Rescue Script (Use This Word-for-Word)
When someone says “That doesn’t match our numbers,” pause for 4 seconds, then say:
“That’s a different figure — can I ask what data source that’s from? … Both numbers are measuring [same thing]. The difference is [methodology/period/scope]. For the decision we’re making today, the relevant figure is [X] because [specific reason].”
This covers 80% of data contradictions. The other 20% — hostile challenges, genuine errors, interpretation disputes — need the full framework below.
⭐ Turn Data Contradictions Into Credibility Wins
The Executive Q&A Handling System gives you the complete data defence framework — including the Parallel Truth scripts, question forecasting to predict data challenges BEFORE the meeting, and the Headline → Reason → Proof → Close response structure for handling any challenge in 20-45 seconds.
Your data defence toolkit:
- Data contradiction response scripts — Parallel Truth, Acknowledge-and-Verify, Methodology Bridge
- Question forecasting framework — predict who will challenge which numbers before the meeting
- 7 question type handlers — including the “conflicting data” and “credibility challenge” types
- Headline → Reason → Proof → Close response structure — answer any data challenge in 20-45 seconds
Get the Executive Q&A Handling System → £39
Built from 24 years of data-heavy executive presentations in banking — where conflicting numbers were a weekly reality, not an edge case.
The 3-Step Recovery: Pause → Identify → Redirect
When someone contradicts your data in a presentation, your nervous system fires before your brain catches up. Heart rate spikes. Palms sweat. The instinct is immediate defence. Override that instinct with this 3-step sequence:
Step 1: Pause (4 seconds). Take a breath. Look at the person who challenged you. Nod once. These 4 seconds do three things: they signal composure to the room, they give your brain time to move from defensive reaction to analytical thinking, and they create a moment of gravity that says “I’m taking this seriously.” The audience reads this as confidence, not hesitation.
Step 2: Identify the difference (15-30 seconds). Ask one diagnostic question: “Can I ask what data source that’s based on?” or “Is that using the same reporting period?” or “Does that include [specific scope element]?” You’re not interrogating them — you’re collaborating on finding the discrepancy. The answer will almost always reveal a methodology, time period, or scope difference. Now you have the explanation.
Step 3: Redirect to the decision (15-20 seconds). “That explains the difference — [their approach] measures X, mine measures Y. For the decision we’re making today, the relevant figure is [X/Y] because [specific reason].” This closes the loop and moves the room forward. You’ve acknowledged the discrepancy, explained it, and focused everyone on what matters.
Total time: 30-50 seconds. In that window, you’ve transformed a potential credibility crisis into a demonstration of analytical depth and composure. The room remembers how you handled the challenge, not the challenge itself.
For a complete system on predicting presentation questions before you walk in, including data challenges, see the Question Map framework.
The 4 Worst Responses (And Why Smart People Default to Them)
“No, that’s not correct.” The most natural response — and the most damaging. You’ve just called a colleague wrong in front of the room. Even if you’re right, the dynamic has shifted from presentation to confrontation. The room is now watching a conflict, not evaluating a recommendation. And if the other person outranks you, you’ve just created an enemy with more organisational power.
“Let me check and get back to you.” This sounds professional but it’s a credibility withdrawal. The room hears: “I’m not confident enough in my own numbers to defend them.” In an executive meeting, deferring on your own data signals that you haven’t done sufficient preparation — even if the real issue is a legitimate methodological difference you could explain in 20 seconds.
“My methodology is [detailed technical explanation].” Over-explaining in the moment makes you sound defensive. A 90-second methodology defence while 12 executives wait is 90 seconds too long. The room doesn’t want a statistics lecture — they want to know which number to use and why. Save the methodology for the follow-up if someone specifically asks.
“That’s interesting — we should compare notes after the meeting.” This sounds collaborative but it punts the issue. The room still doesn’t know which number is right. Worse, it creates an unresolved tension that hangs over the rest of your presentation. Every subsequent data point will be received with slightly more scepticism because the first one was unresolved.
The response frameworks in the Executive Q&A Handling System (£39) give you the Headline → Reason → Proof → Close structure for handling data challenges in 20-45 seconds — so you never default to the 4 worst responses under pressure.
How to Prevent Data Contradictions Before They Happen
The best data contradiction is the one that never occurs. Three pre-meeting steps eliminate 80% of in-room challenges:
Step 1: Source your data from the audience’s preferred system. If the CFO uses Dashboard X, pull your data from Dashboard X. If the risk committee trusts Model Y, use Model Y. When your data comes from the same source the room trusts, contradictions become nearly impossible. If you need to use a different source, flag it on the slide: “Source: Internal Ratings Model (differs from standardised approach by approximately 1.5-2%).” Pre-empting the discrepancy removes its power.
Step 2: Pre-meet with the most likely challenger. Who in the room has their own data? The CFO? The head of risk? The regional lead with different divisional numbers? Meet them 48 hours before the presentation. “I’m presenting Q3 performance on Thursday. My figures show 4.1% using the IRB model. I understand your team tracks this differently — can you share your latest figure so I can address any discrepancy proactively?” This conversation either aligns your numbers or prepares you for the exact discrepancy that will arise.
Step 3: Build the discrepancy into your deck. If you know different numbers exist, address it in a slide: “Note: Credit risk team reports 2.3% using the standardised approach. This analysis uses IRB methodology, which captures concentration risk. The 1.8% difference reflects [specific factor].” When the challenge comes — and it will — you point to the slide. “I anticipated this — it’s addressed on slide 8.” Now you look thorough rather than caught off guard.
These steps mirror the question forecasting approach from the Q&A preparation system — applied specifically to data-driven presentations.
📋 The 5 Most Common Causes of Conflicting Numbers in Executive Meetings
Before you panic, diagnose. In 24 years of data-heavy executive presentations, nearly every contradiction traced back to one of these five root causes:
| Root cause | What’s happening | Your diagnostic question |
| Different time period | Your Q3 ≠ their Q3 (calendar vs fiscal, rolling vs fixed) | “What reporting period does that cover?” |
| Different methodology | IRB vs standardised, gross vs net, model vs benchmark | “What model or calculation is that based on?” |
| Different scope | Company-wide vs division, all accounts vs top 50 | “Does that include [specific segment]?” |
| Different data source | CRM vs finance system, manual export vs automated feed | “Which system did you pull that from?” |
| Different assumptions | With/without outliers, adjusted vs unadjusted, including FX | “Does that adjust for [specific variable]?” |
The pattern: ONE diagnostic question per cause. Identify the root cause, and the contradiction resolves itself. → The Executive Q&A Handling System (£39) includes question forecasting scripts to predict which of these five causes will surface — before the meeting starts.
When You ARE Wrong: The 15-Second Correction That Builds Trust
Sometimes the contradiction isn’t about methodology. Sometimes your number is genuinely wrong. A transposition error. An outdated data pull. A formula mistake in the spreadsheet.
When this happens, speed and honesty win everything.
The script: “You’re right — thank you for catching that. The correct figure is [X]. That changes [specific implication] but doesn’t change the recommendation because [reason]. I’ll update the deck and circulate the corrected version by end of day.”
This takes 15 seconds. In that window, you’ve done four things: acknowledged the error without drama, corrected the record, assessed the impact on your recommendation, and committed to a follow-up action. The room sees someone who handles errors with the same composure as they handle everything else.
What kills credibility isn’t being wrong. It’s being wrong and pretending you’re not. Every executive in the room has presented incorrect data at some point. They recognise and respect the clean correction. What they don’t respect: arguing for 5 minutes and then quietly acknowledging the error. That costs you far more than the original mistake.
The Executive Q&A Handling System (£39) includes scripts for every Q&A scenario — including the “I was wrong” correction, the “both numbers are right” Parallel Truth, and the “I need to verify” bridge. Pre-written responses for the moments where thinking on your feet isn’t enough.
Is This Right For You?
✓ This is for you if:
- You present data to executive audiences and live in fear of someone producing different numbers
- You’ve been publicly contradicted on data before and want a framework for handling it next time
- You want to forecast data challenges BEFORE the meeting and prepare specific responses
✗ This is NOT for you if:
- Your presentations don’t involve data or quantitative analysis
- You’re looking for help with general presentation structure (see the Executive Slide System for that)
Built from 24 years of credit committees, risk reviews, and data disputes at JPMorgan, PwC, RBS, and Commerzbank. Question forecasting. Response scripts. 7 question type handlers. Instant download.
Frequently Asked Questions
What if the person contradicting me is more senior?
The Parallel Truth Framework works even better when the challenger outranks you. By saying “Both numbers are accurate — they measure different things,” you’re not contradicting someone senior. You’re validating their figure while explaining yours. The diagnostic question (“What data source is that from?”) is also easier to ask upward — it reads as deference and genuine curiosity rather than challenge. The key: never say “my numbers are right.” Say “the difference is [specific methodology].” This lets the senior person maintain authority while you maintain credibility.
How do I prepare if I don’t know who might challenge me?
Three rules. First, know your methodology cold — one sentence, no jargon: “This uses the IRB model for the Q3 period, including concentration risk.” Second, prepare the top 3 alternative methodologies someone might use and the approximate difference each would produce. Third, build a slide that pre-empts the most likely discrepancy. If you can’t identify a specific challenger, you can still identify the most common analytical differences in your field. This general preparation handles 80% of unexpected challenges.
What if the contradiction is about my interpretation, not my data?
Different challenge, similar principle. When someone disputes your interpretation (“The data doesn’t support that conclusion”), the response is: “That’s a fair reading. My interpretation is based on [specific assumption]. If we change that assumption to [theirs], the data could support [their conclusion]. For this decision, I’ve used [your assumption] because [specific reason].” You’re not arguing about who’s right — you’re showing that the interpretation depends on an assumption, and explaining why yours is the right one for this context.
Should I ever concede a data point to avoid conflict?
Only if the concession doesn’t change your recommendation. If someone challenges a minor data point and your recommendation stands regardless, say: “Fair point — even using your figure, the recommendation holds because [reason].” This is powerful. It shows the room that your conclusion is robust — it survives a data challenge. But never concede a data point that undermines your core recommendation just to avoid conflict. That’s not diplomacy — that’s abandoning your analysis. If the data matters, defend it professionally. If it doesn’t, concede it gracefully.
📬 The Winning Edge — Weekly Newsletter
One executive presentation insight per week. Q&A recovery, data defence, and the frameworks that turn difficult moments into career-building ones. No filler.
Read next: If you’re presenting a deck you didn’t build (where data challenges are even harder to handle), read Presenting When You’ve Inherited Someone Else’s Deck — the 90-minute Transplant Method.
Read next: If your data presentation goes to your boss’s boss, read The Skip-Level Presentation: What Changes When You Present to Your Boss’s Boss.
Your next data-heavy presentation is on the calendar. Someone in that room has different numbers. Get the framework that turns “that doesn’t match our figures” into the moment the room trusts you most.
About the Author
Mary Beth Hazeldine is the Owner & Managing Director of Winning Presentations. With 24 years of corporate banking experience at JPMorgan Chase, PwC, Royal Bank of Scotland, and Commerzbank, she has delivered high-stakes presentations in boardrooms across three continents.
A qualified clinical hypnotherapist and NLP practitioner, Mary Beth combines executive communication expertise with evidence-based techniques for managing presentation anxiety. She has trained thousands of executives and supported presentations for high-stakes funding rounds and approvals.