The Finance Presentation: Pre-Read, Deck and Speaking to It

A presentation is three documents, not one. The pre-read carries the whole argument and lands before the meeting. The deck carries only what the room has to look at together. The spoken layer carries the reasoning and the limits around it. Building one thing and using it for all three is why finance meetings end without a decision.

The work is already done. The sources are logged, the assumptions are written down, the review pass is finished and somebody has read it all back to the analyst. Now twenty minutes have been booked, six people are coming, and none of that finished work does anything at all until it gets out of one head and into six. The delivery is not a smaller version of the analysis; it is a separate job of construction with its own rules, and doing it badly wastes every hour that came before it.

One assignment runs through this guide, the same one that runs through the rest of the sequence. The Kavery research desk, inside the invented firm Kavery Capital Services Private Limited, was asked a question about Meenakshi Tubes Private Limited, an invented maker of steel tubes. Sharada Iyer produced the work, Prakash Nadar reviewed it, and Latha Menon chaired the meeting where it was settled. The analysis was closed before the meeting was booked. The subject is incidental: put a bond, a fund or a single machine in place of the tube maker and not one sentence has to change.

Whether the action put to that meeting was the right one is a view about a business. How the document carrying it is built, and how somebody speaks to it, is a craft that transfers unchanged to any subject.

Why can a room not simply be handed the analysis?

Everything below rests on one plain fact about people, and it is worth stating before any of the craft: nobody can read and listen at the same time. Reading and listening compete for the same attention, and reading wins the competition. Reading is faster, and it runs at a speed the reader controls.

Ordinary life shows this. A printed menu handed over while the specials are being described does the same thing every time. The person goes quiet, their eyes drop, and thirty seconds later they ask a question that has just been answered. The person was not being rude. The person was reading. The menu was in front of them and the voice was not, and the menu won.

A slide carrying eight lines of text is a menu. The room reads it in about twelve seconds. Reading eight lines is faster than saying them aloud, so the room finishes and then waits while the presenter catches up. Anything said inside those twelve seconds lands on nobody. The gap in speed is why the argument cannot simply be typed onto the screen and read aloud, however complete and however correct that argument is.

The way out is not to say less. The way out is to split the material by how it is best received. Whatever the room has to read goes into a document they read on their own time. Whatever the room has to look at together goes on the screen. Whatever the room has to hear from a person comes out of the presenter's mouth. Three receptions, three carriers, and one split that every rule of delivery comes out of.

What are the three documents, and what does each one do?

Three carriers, three jobs, and each one does work the other two cannot. Here they are in the order the room meets them.

The pre-readThe full written argument, circulated before a meeting so that the room arrives already informed. is the whole argument, written out, sent ahead. The pre-read is a document in the ordinary sense: it has sentences, it has a table or two, and somebody can sit with it for ten minutes and come out knowing what the desk found and what the desk wants. A pre-read exists so that the meeting can start at minute one rather than at minute nine.

The presentation deckThe slides shown in the room, carrying only the material that has to be looked at by everybody at the same moment. is not a shorter pre-read. The deck carries only what the room has to look at together, at the same moment, with somebody pointing at it. The set is much narrower than most people expect, and on most decks it turns out to be the pictures and very little else.

The spoken layer is what comes out of the presenter's mouth. The spoken layer carries the reasoning, the judgement calls, the answers to the questions people ask, and the limits around what is being put forward. Only the spoken layer can respond to the room, so the material most likely to be argued with belongs there rather than on a slide.

Think of a school parents meeting. The report card goes home in the bag beforehand, and that is the pre-read: the parent who reads it arrives with questions instead of surprise. Everybody has to be looking at the same thing while it is discussed, so the chart of class averages on the wall is the deck. And the teacher saying which two subjects she would watch, and why she might be wrong about one of them, is the spoken layer. Three carriers, one meeting, and none of them substitutes for another.

What Is a Pre-Read, and How Far Ahead Does It Have to Land?

The pre-read in this assignment was the memo, laid out over three sheets and sent to the six people attending. Nothing was rewritten for it. The memo already ran to fourteen hundred words with the answer and the request in the first sixty, so it was already the document a reader could work through alone, and reformatting it into slides would have destroyed the only version that worked on its own.

The fourteen hundred words sat across the three sheets like this, with the two exhibits placed where the argument needed them.

Twenty four hours is the interval that did the work, and both directions away from it fail. Send it a week ahead and nobody reads it. A week is long enough for the meeting to feel like something happening to a future person. Send it thirty minutes ahead and it is not a pre-read at all, it is an attachment: nobody reads three sheets standing in a corridor. Twenty four hours is long enough to be read at somebody's own desk and short enough that what they read is still in their head when they sit down.

Four of the six had read it. The number four is worth sitting with. The instinct on hearing it is disappointment, and disappointment leads straight to the wrong repair. Two thirds is a normal return on a pre-read, not a failure of one. The right response is to build the deck so that the two who did not read it can still follow the meeting. The wrong response, and it is the common one, is to conclude that pre-reads do not work and to move the entire argument onto the slides. Moving the argument onto the slides is precisely what stops the room listening to anybody.

What Goes on a Presentation Deck, and What Never Does?

Now the deck, and this is where most of a working week gets spent for the least return. The instinct is to treat the deck as a compressed pre-read: take the argument, shorten every sentence, put the shortened sentences on slides. The instinct produces a document that is neither a pre-read nor a deck, and serves as neither.

The deck carries two kinds of thing and nothing else. The first is anything the room has to look at together, in the same second, with somebody pointing at part of it. A picture is the obvious case: six people cannot discuss the shape of a line unless all six are looking at the same line at the same moment. A four row table is the same case in a different form. The second is anything six people have to copy down identically. In practice that is the request being made, and the one sentence that has to be repeated to somebody who was not in the room.

The slide test, and the one thing it lets through

The slide testA single question used to decide whether material belongs on a screen: if it could be said out loud instead, it is not a slide. is one question: could this be said out loud instead? If it could, it is not a slide. Anything that could be said belongs in the spoken layer, and if it also has to be on the record it goes into the pre-read, where it already is. Anything on a screen that could simply have been said is competing with the presenter for the attention of the room, and the presenter loses that competition every time.

The request can obviously be said out loud, so a test applied without any exception deletes the request as well. So the test carries one carve-out, and naming it is what makes the test usable rather than a slogan. Material stays on the screen when six people have to write down the same words. A number said aloud gets recorded six slightly different ways; a number on a screen gets recorded once. Identical wording is a real reason for something to be looked at together, and it is the only reason a text-only slide survives.

Sharada Iyer drafted fourteen slides. The test cut five of them and left nine. Nothing was lost. Everything cut was either said out loud in the room or already sitting in the pre-read that four of the six had read.

A Slide Deck Is Not a Document, So What Breaks When It Is Used as One?

There is a second reason decks fill up with text, and it is not laziness. The reason is that the slide deckThe same object as a presentation deck; the two names are used interchangeably for the slides shown in a room. gets emailed afterwards to people who were not in the room, and it gets filed as the record of what was said. Both of those are real needs. Neither of them is a reason to change the deck, and the whole problem comes from thinking they are.

A deck that has been rewritten so it makes sense to an absent reader is a document set in a very wide typeface with pictures scattered through it. A document has paragraphs and a deck has fragments, so the hybrid is worse than a document at being read. Everything on it is now text, so the hybrid is worse than a deck at being shown. Optimising one artefact for two incompatible readers produces something that serves neither, and the fix is not a better deck but a second file.

The fix is the pre-read already written. The person who missed the meeting gets three sheets that were built to be read alone, and the file of record is the decision record with the pre-read attached to it. The deck is a set of visual aids for twenty minutes in one room, and once those twenty minutes are over it has finished its work.

How many slides, and what actually decides the number?

The number of slides is set by the length of the slot and never by the amount of material. Twenty minutes were booked, and twenty minutes is what the deck had to be built to fit, whether the working file behind it held twenty one documents or two hundred.

The division settles what is actually on offer. Fourteen slides in twenty minutes is one minute and twenty six seconds each. Nine slides in twenty minutes is two minutes and thirteen seconds each. No meeting spreads its time evenly, so neither of those is a plan, but the second one at least leaves room for a room to interrupt. A slide count that leaves no room for interruption is not a presentation, it is a reading, and the only thing a reading tests is whether the presenter can finish on time.

Verbal Communication: What Does the Spoken Layer Carry That Neither Document Can?

Verbal communicationWhat is said out loud in the room, which carries the reasoning and the limits that the written carriers cannot adapt to a live question. is the layer people prepare least and it is the one carrying the most. The spoken layer holds the reasoning between the numbers, the judgement calls that a table cannot show, every answer to every question, and the limits around what is being put forward. The spoken layer is also the only one that can change while it is being delivered.

A script cannot answer a question, so almost none of the spoken layer is scripted. One part is scripted word for word, and that is the opening ninety seconds. There are two reasons. The first is that the opening is where a presenter is most nervous, and a nervous person without a script either rambles or races. The second matters more: the first ninety seconds decides what the room thinks the meeting is about, and once six people have formed that view the rest of the slot goes on arguing with it rather than with the material.

The scripted opening in this assignment ran to one hundred and ninety words in four parts.

Part four is the one people leave out, and it costs them the rest of the slot. Telling the room where the argument lives releases the presenter from having to prove the argument on the screen. Here the argument lived on the three sheets that went out yesterday. Naming the point at which to interrupt turns a dreaded silence into a scheduled event. An opening that invites interruption at a named point is how a presenter stops being interrupted at every other point.

How to Present Financial Analysis to a Non-Technical Audience: What Actually Changes?

Two of the six in that room did not work with numbers all day. One ran operations and one sat on the committee because a decision of this size needed a second signature. The material has to reach them, and the way most people try to do that is by taking the numbers out. Taking the numbers out is exactly backwards.

Translating for a non-technical reader means dropping the term and keeping the number, and every version that does the opposite has thrown away the only part the audience could have used. A person who does not know what an operating margin is understands perfectly well what is left over from a hundred rupees, and they can compare this year with last year without any help at all. Nobody can do anything with a sentence saying that margins are a little thin.

Here is the translation that went on slide five, in plain languageThe same content stated without the technical term, keeping every number that was in the technical version. and nothing else. For every Rs 100/- of sales, Rs 10.80 is left after running costs, against Rs 12.00 last year. The plain sentence carries the identical content to the technical one. Nobody has been talked down to, no accounting has been explained, and the two people who do not work with numbers now hold the same fact as the four who do.

One warning about the same move used in the other direction. Translating is not softening. If the plain sentence makes the position sound better than the technical one did, that is editing rather than translating. The test is simple and worth running out loud: does the plain version let a listener reach a different conclusion from the technical version? If it does, the sentence goes back.

How to Explain a Financial Recommendation With Appropriate Boundaries: Which Three Parts?

The spoken proposal is the part of the slot where careful work most often becomes an overstated sentence, and the reason has nothing to do with dishonesty. Writing gives time. A claim in the pre-read can be qualified at leisure, read back, and softened where it went further than the evidence. Speaking gives no time at all. Somebody asks a direct question, six people look at the presenter, and the qualification that took two clauses in writing gets dropped because it felt slow to say.

The repair is to fix the shape of the sentence in advance so that the qualification is not something to be remembered and added. A proposed action said out loud has three parts in one sentence: the action, the basis it rests on, and the condition that would change it. Three parts, one breath, and no order other than that one. The basis and the limit have to arrive attached to the action rather than trailing behind it as an afterthought.

The same shape turns up at home without a name for it. Nobody says the car needs replacing. The person says the mechanic has quoted Rs 22,000/- for the clutch, that this is on his estimate rather than a strip down, and that if the gearbox turns out to be involved the number changes. Action, basis, limit. Nobody in that kitchen is misled about how much is known, and nobody is left unable to decide either.

Note what the three parts do not do. The three parts do not hedge the action. The proposal is stated once, plainly, in ordinary words, and the person hearing it can accept it, reject it or send it back. A boundaryThe stated limit of what a piece of analysis supports, said at the same time as the conclusion rather than afterwards. is not a retreat from a claim; it is the edge of the claim, drawn while the claim is being made. A sentence with no edges is not braver than one with edges, only less finished.

Pre-Read vs Presentation Deck: Where Exactly Do the Two Part?

The two carriers get confused with each other constantly, and the confusion is understandable. Both are files, both go to the same six people and both are built from the same finished work. Set them side by side and the differences are not matters of degree. The two carriers differ on what they carry, when they land, how they are received and what happens to them afterwards.

Read the table one more way. Every row on the left is about a person alone with a document, and every row on the right is about six people in a room with a screen. Reading alone and watching together are two different acts of reception, and no amount of careful writing makes one file do both of them.

What did this deck and this meeting actually look like?

Nine slides for a twenty minute slot, built for a room in which two of the six had not read anything. Here is the whole deck, with the reason each slide earned a screen and the minutes it actually took.

Slide six was the one that mattered. Slide six showed what happens to the reported position when the December order is taken out. The single December order supplied Rs 2,96,80,000 of the Rs 21,20,00,000 reported, 14 per cent of the year, and it sits inside the third quarter, so the quarterly slide came two slides earlier. With it removed, the plain sentence becomes Rs 5.00 left from every Rs 100/- of sales, against the Rs 10.80 the year actually reported.

Now look again at the last row. Slides seven, eight and nine were never reached, and slide nine was the one carrying the request and the review date. The meeting still produced a decision, and the only reason it could is that the request was also on slide one. Putting the answer and the request first is Barbara Minto's ordering from The Pyramid Principle in 1978. In a deck the ordering is not merely polite; it is what makes a deck survive never being finished. A deck that builds towards its request at the end is a deck that hands nothing to a room that runs out of time, and rooms run out of time.

What does it mean when the room spends all its time on one slide?

Twelve of the twenty minutes went on slide six. Nothing before it took more than two minutes, three slides were never shown, and the presenter walked out having delivered two thirds of the deck.

The instinct is that this went badly. It did not. Time in a meeting does not spread across the material; it pools around the one thing the answer actually depends on, and where it pools is the most useful reading available all week. The room stopped on the scenario because the scenario is where the answer is fragile, and six people had independently worked out that the whole conclusion rests on whether one order repeats. Twelve minutes on one slide is not a failure of preparation. The room was doing precisely the job it was assembled to do.

There is a practical consequence, and it changes how the next deck is built. If the minutes are going to pool anyway, the slide they will pool on goes early enough that pooling does not cost the request. Slide six with fourteen minutes left in front of it survives twelve minutes of argument. Slide six with six minutes left in front of it does not, and neither does anything behind it.

What goes wrong when the deck is built as a document?

How does a lender, an analyst or somebody working alone use any of this?

A credit committee at a lender runs on exactly this split, whether or not anybody names it. The paper goes out ahead, the exhibits go on the screen, and the committee spends its time on the one assumption it does not believe. A committee member who has read the paper asks a question at minute three; one who has not asks the presenter to go back to the table. Both are normal, and the deck has to serve both without turning into the paper.

An analyst putting work in front of a portfolio manager has the same three carriers compressed into a smaller room. The note goes over the previous evening, the two charts go on a screen, and the spoken layer is almost entirely answers. In a two person conversation the qualification is the easiest thing in the world to leave out, so the three-part sentence matters more here, not less.

For somebody working alone, with no committee, no reviewer and nobody to present to, none of this becomes irrelevant. The pre-read becomes the document written for oneself and read back cold the next morning. Reading it back cold catches more than any amount of staring at a screen. The slide test becomes a filter on what goes in front of a client, a partner or a bank manager. And the three-part sentence is how an intended action is stated without quietly promising more knowledge than the work supports. The carriers exist because of how people receive things, and a person receiving their own work a day later is subject to the same limits as a room of six.

The household version is not a metaphor. Before a large purchase, somebody in the house circulates the numbers the evening before, everybody looks at the same two figures together at the kitchen table, and one person says what they think should happen, on what basis, and what would change their mind. The kitchen conversation goes well or badly for exactly the reasons a committee meeting does.

The Kavery research desk, Kavery Capital Services Private Limited, Meenakshi Tubes Private Limited, Sharada Iyer, Prakash Nadar and Latha Menon are invented.
Educational material. Not advice on any investment, tax, budget or market position.