Tag: presenting data to executives

In 2003 I sat on a credit committee at one of the banks where I was working. The committee was seven people, only two of whom could read a cash-flow model without help; the rest were senior commercial people who decided with judgement, not with spreadsheets. A relationship director came in to present a lending proposal he believed was a formality. He had eleven data slides — leverage ratios, sector comparables, a sensitivity table with nine scenarios, a debt-service waterfall, and a covenant-headroom chart with three overlaid lines. He talked us through all eleven. At minute fourteen the chair stopped him, flipped back through the printed pack, circled three numbers in red pen on three different slides, and said: “Come back when you can tell us, on one page, what these mean and what you want us to approve.” The proposal was sound. The data was sound. It was deferred for two weeks anyway, because the committee had been handed a dataset and asked to find the decision inside it themselves.

(This article was created with AI assistance; all stories and insights are based on 35 years of real client work.)

That deferral taught me the rule I have spent the years since teaching to senior professionals who present data to boards: a non-technical board does not get more confident as you give it more data. It gets less confident, because every additional chart is one more piece of interpretation you have pushed onto people who are not equipped — or paid — to do it in the room. This piece walks through the structural method that reverses the effect: the Decision-First Data Slide, the eight-second glance test that tells you whether a slide is board-ready or analyst-ready, and where the eleven charts you cut actually belong. The method is named, testable, and works whether your board is a credit committee, a main board, a charity trustee meeting, or an investment committee.

Why more data makes a non-technical board decide less

The instinct that drives the eleven-slide deck is reasonable and almost always wrong. The presenter reasons: this is an important decision, the board should see the rigour behind it, so I will show the analysis. The flaw is in the word “show.” A non-technical director cannot absorb a sensitivity table with nine scenarios in the forty seconds the slide is on screen. What the director actually does is glance at it, register that it looks thorough, fail to extract a conclusion, and feel a small flicker of unease — because they have just been shown something important that they could not read. That unease does not register as “I do not understand this chart.” It registers as “I am not sure about this proposal.” The presenter has converted analytical rigour into decision hesitancy, which is the exact opposite of the intention.

This is the mechanism behind the paradox. Every chart that a non-technical audience cannot interpret in the moment adds cognitive load without adding confidence, and cognitive load that does not resolve into a conclusion is experienced as risk. The board is not thinking “the leverage ratio is concerning.” It is thinking “there is a lot here I am being asked to take on trust, and I do not yet know what I am being asked to approve.” A board that does not know what it is being asked to approve does the safest available thing, which is to ask for more work and defer. The eleven slides did not fail because they were wrong. They failed because they asked the wrong people to do the interpretation. The broader principles of presenting data to executive audiences all reduce to this single idea: the interpreting is the presenter’s job, and a data slide that outsources it back to the board has failed before anyone speaks.

The reframe that fixes it is uncomfortable for people who built their careers on analytical depth. You are not presenting your analysis. You are presenting the decision your analysis produced, supported by the minimum evidence a non-technical person needs to trust it. The depth still exists — it lives in the appendix, in the model, in the answers you give when challenged — but it is not the spine of the presentation. The spine is the conclusion and the one chart that makes the conclusion visible. Everything else is available on request, not on screen by default.

The Decision-First Data Slide: five parts

The Decision-First Data Slide is a single slide built to five rules. It is not a style preference; each rule removes one of the ways a data slide pushes interpretation back onto the board. A slide that follows all five carries one decision and makes the supporting number visible in a single glance. A slide that breaks any of them starts to read as an analyst’s working paper rather than a board input.

The five parts are these. One, the headline states the conclusion, not the topic. “Revenue is on track to miss the full-year plan by nine percent unless we act on pricing this quarter” is a conclusion. “Q3 Revenue Performance” is a topic, and a topic headline forces the board to read the chart to find out what they are meant to feel about it. Two, one chart carries the slide — not a dashboard, not four panels, one chart that the conclusion depends on. Three, the decisive number is annotated directly on the chart, so the eye lands on it without searching: a callout arrow, a bold data label, a shaded gap between the actual line and the plan line. Four, the comparison is built in. A number with nothing to compare it to is not decision information; the board needs to see the actual against the plan, the target, or the prior period, on the same chart. Five, the source sits on one line at the bottom — quietly, in small type — so the provenance question is pre-answered without cluttering the decision.

What makes the five parts work together is that they each protect the board from a specific failure. The conclusion headline removes the interpreting. The single chart removes the searching. The annotation removes the hunting for the relevant number. The built-in comparison removes the “is that good or bad?” gap. The source line removes the provenance doubt before it is voiced. None of them is about making the slide prettier. Each is about removing a task the board would otherwise have to do in the room, and the cumulative effect of removing those five tasks is a slide a non-technical director can act on in seconds rather than puzzle over for a minute.

The eight-second glance test

The Decision-First Data Slide needs a test that tells you, before the meeting, whether a slide is board-ready or analyst-ready. The test is mechanical and it is the single most useful thing in this article. Take the slide. Hand it to a colleague who does not work in your function — someone from HR, operations, legal, anyone who was not in the analysis. Give them eight seconds with it, then take it away. Ask two questions: which way is the number pointing, and what are we being asked to decide? If they can answer both in one sentence, the slide is board-ready. If they hesitate, hunt back over the slide in their memory, or say “well, it depends what you mean,” the slide is analyst-ready and the board will do exactly what your colleague just did — except the board will not tell you, it will simply defer.

Eight seconds is not arbitrary. It is roughly the attention a non-technical director gives a data slide before they either lock onto its conclusion or start half-listening to your voice while their eyes drift. If the slide has not delivered its conclusion in that window, you have lost the slide; the rest of your forty seconds of talking is spent against a director who is now reading the chart instead of listening to you. The eight-second test simulates the real conditions of the room. It is deliberately unfair, because the room is unfair: nobody on a board studies your slide the way you studied your model.

The test also diagnoses which of the five parts has failed. If your colleague cannot tell which way the number points, the annotation or the comparison is missing. If they can read the number but cannot say what is being decided, the headline is a topic, not a conclusion. If their eyes visibly dart around the slide, you have more than one chart competing for the decisive glance. The test does not just tell you the slide is wrong; it tells you which rule to fix. Run it on every data slide in your next board pack and you will cut the pack down before the board ever sees it. The visualisation choices that survive an executive glance are the ones engineered to pass this test, not the ones that look most impressive on a wide monitor in your own office.

The conclusion headline that does the interpreting

Of the five parts, the conclusion headline does the most work and is the one presenters resist most. A conclusion headline commits you. “Q3 Revenue Performance” commits to nothing; if the board reacts badly you can retreat into “well, I was just showing the data.” “Revenue will miss the full-year plan by nine percent without a pricing decision this quarter” commits you to a position the board can challenge. That exposure is precisely why it works. A board engages with a position. It cannot engage with a topic, because a topic gives it nothing to agree or disagree with, so it falls back on the only response a topic allows: more questions, more requests, more delay.

The discipline of writing the headline is the discipline of finishing your own thinking before the meeting. If you cannot write the conclusion in one sentence, you have not actually concluded — you have assembled evidence and hoped the board would conclude for you. Writing the headline forces the analytical work to finish at your desk, where it is cheap to be wrong, rather than in the boardroom, where it is expensive. The headline is also the line the board will quote back when they brief the people who were not there. If you do not write it, someone else will write a worse one for you, in the corridor afterwards, with your name attached.

There is a craft to making the headline land without overclaiming. It states the conclusion as a process finding, not a guaranteed outcome: “on current trajectory, revenue misses plan by nine percent” rather than “revenue will collapse.” It carries the decision implication: “without a pricing decision this quarter.” And it stays in the board’s language, not the analyst’s — “miss the plan” rather than “negative variance to budget at the EBITDA line.” A headline a director can repeat verbatim to another director is a headline that has done its job, because the decision now travels without you in the room. Turning a dataset into a narrative the board can carry is the same skill applied across the whole deck rather than a single slide.

One chart, not a dashboard, on a decision slide

In 2019 I ran a preparation session with a finance director rehearsing a board paper on a pricing decision. Her draft opened with a slide carrying four charts — volume, margin, competitor pricing, and customer churn — arranged in a two-by-two grid she was proud of, because each chart was relevant and she had built them all herself. I asked her the eight-second question. She could not, in eight seconds, tell me which chart the decision turned on. We rebuilt the slide to carry one chart: the margin trend, with the actual line and the plan line and the shaded gap between them, the gap annotated with the nine-percent figure, and a headline that stated the conclusion. The other three charts went to the appendix. She presented it the following week. The board approved the pricing change in under twenty minutes, and one director told her afterwards it was “the clearest board paper we’ve had this year.” Same analysis. One chart on the decision slide instead of four.

The reason one chart beats four is not aesthetic. A decision slide has exactly one job: make the conclusion’s supporting evidence visible in a glance. Four charts force the board to choose which one to look at, and while they choose, they are not deciding. The other three charts are not wrong — volume, competitor pricing, and churn all mattered to the analysis — but they are answers to questions the board has not yet asked. Putting them on the decision slide is answering questions out of order, and out-of-order answers read as clutter. They belong where they can be reached the instant a director asks the question they answer, which is the appendix, not the spine.

The test for whether a chart belongs on the decision slide is simple: does the conclusion in the headline depend on this specific chart? If the headline is about a nine-percent margin miss, the margin chart stays and everything else goes. If a director asks “what about churn?” you turn to the churn slide in the appendix, which makes you look prepared rather than cluttered. The discipline of one chart per decision is the discipline of trusting the board to ask for the second chart when they want it — and they will, which is exactly why it should be ready and not on screen.

Where the other ten charts belong

The fear that stops people stripping a deck down is that the depth will look missing — that a one-chart decision slide will read as thin, as though the presenter has not done the work. The opposite is true, but only if the depth is visibly available rather than absent. The eleven charts from the 2003 credit committee were not the problem; their position was. As the spine of the presentation they caused a deferral. As a tabbed appendix, reached the moment a committee member asks a question, they would have signalled exactly the rigour the presenter wanted to convey — without forcing the committee to wade through it to find the decision.

A board paper for a non-technical board therefore has a clear two-tier shape. The front is decision-first: conclusion, the one chart that supports it, the recommendation, the ask. The back is analyst-grade: the sensitivity tables, the comparables, the model assumptions, the scenario range, the methodology. The front is what you present. The back is what you reach for when challenged, and reaching for it confidently is what builds the board’s trust that the front is sound. A presenter who can answer “what happens if rates rise two hundred basis points?” by turning straight to the scenario in the appendix has demonstrated more rigour than a presenter who put all nine scenarios on the front slide and let the board drown in them.

This two-tier shape also solves the political problem that makes people over-show data in the first place: the fear of looking unprepared in front of the one technical person on the board. There is often a finance director, an audit-committee chair, or an ex-banker who can read the model. The instinct is to build the whole deck for that one person. Do not. Build the front for the non-technical majority who hold the decision, and build the appendix for the one technical reader who will go looking for the detail. The appendix satisfies the specialist without sacrificing the decision-makers, and the specialist, far from feeling short-changed, usually appreciates being able to find the detail in a structured back section rather than scattered across the main slides.

Why this matters more for senior presenters than for analysts

An analyst presenting to other analysts can lead with the working, because the audience reads working fluently and judges the analyst on the rigour of the method. A senior professional presenting to a non-technical board is judged on something different: whether they can take a complex analysis and extract from it a decision the board can act on. That extraction is the senior skill. A board does not promote the person who can build the most detailed sensitivity table; it trusts the person who can stand up, state what the table means in one sentence, and stand behind it. The Decision-First Data Slide is the visible form of that skill. It says, before you have spoken a word: I have done the interpreting, here is the decision, the evidence is one glance away, and the depth is in the back if you want it.

The cost of getting this wrong compounds across a career in a way that is easy to miss. A senior leader who repeatedly hands boards analyst-grade decks earns a quiet reputation as someone whose work needs translating before it can be used — capable, but not yet board-level. A senior leader who repeatedly hands boards decision-first slides earns the opposite reputation: the person whose papers can be acted on as presented. Over a dozen board cycles, those two reputations diverge sharply, and they diverge on a skill that has almost nothing to do with the quality of the underlying analysis and almost everything to do with who is made to do the interpreting in the room.

One thing to do before your next board deck

Take the data slide you are most proud of in your next board pack — the one with the most analysis on it. Write its conclusion in one sentence at the top: what the number means and what the board should do about it. Strip the slide to the single chart that conclusion depends on, annotate the decisive number directly on the chart, and build in the comparison so the board can see whether the number is good or bad without being told. Move every other chart to a clearly tabbed appendix. Then hand the slide to someone who was not in the analysis, give them eight seconds, and ask which way the number points and what is being decided. If they answer in one sentence, you are ready. If they hesitate, you have just found the slide your board would have deferred — with two weeks left to fix it instead of finding out in the room.

Frequently asked questions

Won’t a non-technical board think I’m oversimplifying or hiding something if I only show one chart?

This is the fear that keeps the eleven-slide deck alive, and the two-tier structure is what resolves it. You are not hiding the depth; you are positioning it. The decision slide carries the conclusion and the one chart it depends on, and a clearly tabbed appendix carries the full analysis — the sensitivity tables, the comparables, the model assumptions. When a director asks for more, you turn straight to it, which demonstrates rigour rather than concealing it. Boards do not read “one chart” as hiding; they read it as a presenter who has done the interpreting and knows which evidence the decision actually rests on. The presenters who get accused of oversimplifying are the ones who strip the depth out entirely. Keep all of it — just move ten charts to the back and leave the one the decision turns on at the front.

What if my board genuinely is technical — mostly finance people who want the detail?

Then your “eight-second colleague” should be drawn from that population, and the conclusion headline still applies even when the audience can read the working. A technical board reads charts faster, but it still decides on conclusions, and it still defers when it cannot tell what it is being asked to approve. The difference is where you set the line between front and appendix: a technical board may want two or three charts on the decision slide rather than one, and a richer set of comparables on the front. But the conclusion headline, the built-in comparison, and the annotated decisive number are not concessions to non-technical audiences — they are how any board, technical or not, locks onto a decision quickly. Run the glance test with a technical colleague and calibrate to what they can extract in the time you actually have.

How long does it take to rebuild a deck this way, and is it worth it under time pressure?

Rebuilding an existing deck to the Decision-First structure usually takes a couple of hours the first time and far less once the pattern is familiar, because most of the work is moving charts to an appendix and writing conclusion headlines you should have written anyway. Under genuine time pressure, the single highest-return move is the headline: spend ten minutes writing a one-sentence conclusion at the top of each data slide, even if you change nothing else. That alone shifts the board from interpreting to deciding. The full rebuild is worth it for any high-stakes board paper where a deferral costs you weeks; for a routine update, the headline pass is often enough. The point is that the structural work is cheap relative to the cost of a deferral, and the headline is the cheapest, highest-leverage part of it.

What’s the most common mistake senior presenters make with board data slides?

Leading with the topic instead of the conclusion. The topic headline — “Q3 Performance,” “Market Overview,” “Pricing Analysis” — feels safe because it does not commit, but the lack of commitment is exactly what makes the board work to find the point. The second most common mistake is the dashboard reflex: putting four or six charts on a slide because each one is relevant, without noticing that relevance is not the test. The test is whether the conclusion depends on the chart. A relevant chart that the conclusion does not depend on belongs in the appendix. Fix those two habits — conclusion headlines and one chart per decision slide — and most of the board-data problem disappears, because together they move the interpreting from the board back to where it belongs, which is with you, before the meeting.

The Winning Edge — weekly newsletter

The Winning Edge is a weekly (Thursday) newsletter for senior professionals who present to boards and executive committees. One short email a week on the structural moves that separate the data decks boards act on from the ones they defer. Subscribe to The Winning Edge →

For the full set of skills behind a board-ready data presentation — slides, storytelling, confidence, and delivery — the seven-product Complete Presenter bundle brings them together as a single resource — £99 for everything, lifetime access.

About the author

Mary Beth Hazeldine is Owner & Managing Director of Winning Presentations Ltd. With 24 years of corporate banking experience at JPMorgan Chase, PwC, Royal Bank of Scotland, and Commerzbank, she advises senior professionals across financial services, insurance, consulting, and technology on the structural moves that turn dense analysis into decisions boards can act on.