Choose a Film Deck Design From Real Pages, Not a Mood Board Alone
Choose a Film Deck Design From Real Pages, Not a Mood Board Alone
A mood board is a vocabulary. It tells you whether a designer is thinking in grain and distance or in clean geometry, which is worth knowing. What it can't tell you is whether that vocabulary survives contact with your pages, because a mood board has no difficult page in it. It's all ideal content: one beautiful still, a fragment of type, no reversal to explain, no financing line that has to be found in eight seconds.
A pitch deck has the other kind of content. So the comparison worth running is not which board you like. It's which system, given the same three awkward pages, you can still read.
Pick three pages that do different jobs. A synopsis with narrative movement in it. An approach page where the visual idea has to be shown rather than described. A status page where someone finds out who's attached and what's still missing. Build all three, in every direction you're considering, from the film's actual copy and equally suitable images. Read them the way they will be read.
That gives you three usable outcomes: choose a direction, commission a specific adjustment, or reject a system that can't carry this film. What it doesn't give you is a verdict on a designer's taste, because taste was never the thing you were buying.
Agree on the comparison before asking for samples
Write down what will be compared before anyone opens a design app. Which pages, how many directions, how many rounds of revision. This is the whole reason the comparison produces an answer instead of a pile of opinions: everyone knows what a finished prototype looks like, so nobody is guessing what you want next.
The word that belongs in that scope is paid. A sample-page comparison is real work — reading your synopsis closely enough to set it, choosing images, making two or three pages that hold together. Asking for it free turns a commissioning decision into an audition, and auditions select for people with spare time. The scope doesn't need to be elaborate. It needs terms both of you agreed to.
The pages you choose matter more than how many. Three cover variations test photography and type pairing three times, and nothing else. Your synopsis has a qualification in it, your approach page has an argument to make, your status page has facts to be found. Those are three different problems, and a direction that solves one of them tells you almost nothing about the others.
If you're also considering a template you bought, or one you're about to buy, it enters the comparison as a third candidate under the same conditions. Templates arrive full of demonstration copy and stock photography, which is ideal content by construction — that's what makes them look finished on the sales page. Replace the copy and the images with your own before you judge anything about the proportions or approve the system. Buying a template settles licensing and time. It doesn't settle fit.
Give every candidate the same reading problem
Identical text in every direction. Not a trimmed version for the designer who said your synopsis "reads better short" — that's how you end up comparing two different films. And not identical images either; equally suitable ones. If one direction gets a photograph that carries the film's central image and the other gets something the designer grabbed at 2 a.m., you've learned which of them had a better night.
Then ask the same three questions of every reader, in the same order, before anyone says a word about color.
Where does the story change direction? Make them point at the sentence. A deck is scanned, not studied, so if the turn is only findable on the second pass, record that. It's the finding.
What does this reference demonstrate? Suppose the approach page cites a photograph of a hallway shot from inside, door ajar. The useful answer is a claim about your film — that doors here are thresholds, the moment you're let in or kept out, not architecture. If readers can only say the image is moody, the page hasn't made its argument.
Which of these people are attached right now, and which roles are open? This is where a status page either works or doesn't. If attached names and open roles sit at the same typographic weight, a reader will tell you what they think is attached, and now you have a room full of people with different pictures of your own project.
Write down the answers close to verbatim, and write down the conditions: who read it, whether they'd read the script, printed or on a laptop, how long they had, what size. A paraphrase recorded without conditions is an anecdote, and anecdotes get repeated for years. Three readers is not a usability study; it's a small set of recorded readings, and small sets are enough to see whether a page is doing its job.
Collect preferences about color and type after that, not before. Not because taste doesn't matter — because taste is the easiest thing in the room to have an opinion about and the least informative thing you can collect about a page.
Test the page that refuses to behave
Every deck has one page that won't sit still. The paragraph with the qualifier in it. The image that's the wrong shape. The status block that grew a row.
So build the test into the prototype. Two deliberate changes.
Add back the sentence you cut for the sample. If you handed over a leaner version of your synopsis "so it fits," you tested the layout against content you don't actually have.
Swap the flattering image for the one you need. Not a worse image — the correct one, which is often taller or busier or has a face in it. A design direction is a set of assumptions about what will be flowing through it. The real image tells you whether those assumptions hold or whether the page was secretly designed around one very cooperative photograph.
Then look at the pages in the place they'll be seen. Not at three hundred percent in the design app, where a page can be beautiful and also four times the size it will ever be. On a laptop at the size a financier opens it, or printed at the size you'd hand across a table. Proportions are a prototype decision tested under those conditions, not a rule that applies to film decks in general — a page that needs one shape for one image may need a different one for the next.
When something goes wrong, check which kind of thing it is. A spacing problem — a caption sitting too close to the line above it, a headline crowding an edge — is a repair. It fixes itself in twenty minutes. A page whose whole proportion was built around one crop is a different situation: when the real image arrives, either the crop eats the point of the picture, or the picture sits whole with dead space on both sides. That page wasn't badly tuned. It was tuned to something you didn't have.
Distinguish a repairable page from the wrong film
Two different judgments arrive wearing the same clothes. "I couldn't find the money line" and "this feels like a comedy and my film is about grief" both come out as vague dissatisfaction in a meeting. They cost wildly different amounts.
The buried status line is the cheap one. Say your status page puts the financing figure at the bottom in small type, under the names, the advisors and the timeline. A reader answers "who's involved" and never sees the number, then asks whether the money is settled — a question the page already answered. Raise the line. Give it weight. Move it above the names. The direction survives, because nothing about the direction depended on where that line sat.
The wrong film is not cheap. If a direction's whole manner — saturated blocks, headline rhythm that sounds like exclamation, wide grinning photography — promises a broad comedy and your film is a restrained grief drama, then legibility is beside the point. Every page can be perfectly readable and still be arguing for a movie you're not making. That's not a page you fix. That's a vocabulary that's wrong for the film.
State the evidence for whichever judgment you're making. Name the page, the element, and the reading consequence: "on the approach page, the headline and the image compete, and readers found the caption first." That's evidence, and it can be acted on. "It doesn't feel like us" is real information about you, and worth saying out loud, but on its own it doesn't tell a designer what to change. And when two people in the room disagree about a color, nothing functional has been established. Ask each of them what they were trying to find on the page and whether they found it. That question usually separates the argument that matters from the one that doesn't.
Approve a direction with one next decision
A direction decision is not "we like B." It's a sentence with three parts: which direction you're taking, what it demonstrated under the tests, and the one change that goes into the next real page. Written down, because in three weeks you'll be looking at a new page and trying to remember what you'd agreed the old one proved.
Then resist the synthesis. The temptation after a comparison is to build a hybrid: this direction's restraint, that one's sidebar, the good idea from the template. Each of those elements was measured inside a system you're now taking apart. Often the element you liked most is the very thing that caused the misreading you just recorded.
And keep the file work out of it. Export settings, access, comprehensive accessibility checks — those are real, they're later, and they're different work. This decision is about whether a design direction carries your film. If a file constraint genuinely decides the issue, say so plainly and treat it as a constraint, not as a design finding.
Two directions, one invented film
The case below is constructed to make the method concrete. The film is invented, the pages don't exist, the readers are hypothetical, and nothing here was built, paid for or tested. What's real is the shape of the findings this comparison tends to produce, and the questions that produce them.
The film is a fictional feature called Uncollected. Nadia is a locksmith. Early on she cuts a spare set of keys for a customer, Mr. Larkin, who says he'll come back for them and doesn't. Two years later he dies, and the set is still in her cabinet, tagged with his address. She drives over to return them. She keeps them. Over the following week she lets herself in twice and changes nothing.
The synopsis page has to add one qualification, because without it the deck promises a search the film doesn't stage: Mr. Larkin isn't a mystery to us. In the first ten minutes we learn the spare key was cut for someone he was hiding from. Nadia never learns it. Same words in both directions.
The approach page argues that this film looks at doors as thresholds — the moment you're let in or kept out — rather than as architecture. Its image is a narrow, portrait-format study of a single door with a hand on the handle. The wide, horizontal hallway photograph the designer first reached for has been retired; it was the cooperative image.
The status page carries the same facts in both: director and producer attached, an actor attached as Nadia, script at third draft, about a third of the budget committed, and cinematographer, editor and the small role of Mr. Larkin still open.
A fictional scope note for this work, which is an example of the shape rather than a rate card: two directions, three pages each, built from the current script copy and the director's portrait study, with one revision round per direction after the first read-through, fee agreed before work starts, and no further pages until a direction is chosen.
Direction A is the spare one. Short headlines, generous white space, one image per page. Its synopsis page carries the qualification as a single line of plain prose directly under the page headline, and in this constructed case readers read it — the paraphrase came back as "she doesn't want to hand them over," which is the film. Its approach page is a full-width horizontal band about a third of the page height, with the headline above it. Its status page is a quiet grid.
Direction B is the dense one. More relations per page, rules between blocks, small labels on each beat of the synopsis: SETUP, COMPLICATION, TURN, AFTER.
Two changes go into both.
The qualification lands without trouble in A. In B, the words are identical, but they sit under the label TURN, and the recorded paraphrase was "so she works out who he was hiding from." The label overrode the sentence. Fixing that means removing the labels, and the labels are the page's entire logic. This is the direction arguing for a shape the film doesn't have: B's synopsis implies an audience learning alongside the protagonist, which is the opposite of the film.
Then the portrait study goes into both approach pages. B absorbs it — the image drops into a column at roughly the width of the text beside it. That fit is real, though the reviewers described the picture as illustrating the note rather than making the argument; the page accommodated the image without letting it carry anything.
A breaks. Cropped into a wide horizontal band, the portrait study keeps the middle of the door and loses the top of the frame and the hand at the handle. Readers reported a picture of a door. The page's whole claim was thresholds, and after the crop it showed a wall. Laid in whole instead, the picture sat centered with empty space to either side, the reading line broke, and two of three readers reached the caption before the headline. The band's proportion was built for a landscape photograph, which is exactly the kind of assumption a prototype exists to find.
The status page produced no difference. Both directions were readable, and a page that passes is information too — not every page in a comparison has to fail for the comparison to be worth running.
The decision: take Direction A. Its one structural failure is on a page, not in an idea. The next real page is the approach page, rebuilt around the portrait study as a column of its own running the height of the page, with the headline and the reference note in the remaining width, re-read afterward to check that the headline lands first. That page will need the image whole; the thing being tested is whether the argument survives when the picture doesn't get cropped into agreement. Nothing about this requires changing the deck's page proportion, and if it did, that would be a larger decision than one page.
What not to do next: import B's margin notes and beat labels into A's synopsis page because they looked organized. They were the mechanism that made readers paraphrase the story the wrong way. A quiet system with one rebuilt page is not a compromise between two directions. It's one direction, tested, with a repair.
Approve Direction A on the strength of what it carried, commission the approach page, and leave the export and accessibility work for the pass that actually looks at files. And don't let this turn into a rehearsal you didn't pay for, or into a rule about what shape every film deck must be. The only general claim here is the one you tested: this system, with this film's difficult pages, can be read.
Frequently asked questions
Why isn't a mood board enough for choosing a film deck design?
A mood board is a vocabulary: it can show whether a designer is thinking in grain and distance or clean geometry, but it has no difficult page in it. It is all ideal content. It cannot show whether that vocabulary survives contact with a synopsis, an approach page, and a status page built from the film's actual copy.
What should be agreed before anyone asks for samples?
Write down which pages, how many directions, and how many revision rounds will be compared. The scope should include payment, because a sample-page comparison is real work. The pages matter more than the number: three synopsis variations test photography and type pairing, not the different problems of synopsis, approach, and status.
How should the test pages be set up?
Build the same three pages in every direction: a synopsis with narrative movement, an approach page that must show the visual idea rather than describe it, and a status page where attached names and open roles can be found. Use identical text and equally suitable images. Then ask the same readers the same questions in the same order before collecting color and type preferences.
What is the difference between a repairable page and a direction wrong for the film?
A repairable page is a spacing or hierarchy problem, such as a financing line buried at the bottom in small type; raising it, giving it weight, or moving it above the names can fix it without changing the direction. A wrong direction is different: if its whole manner promises a broad comedy while the film is a restrained grief drama, every page may be readable and still argue for a movie you are not making. That is a vocabulary wrong for the film, not a page repair.
In the invented Uncollected comparison, why was Direction A chosen?
Direction A's qualification was read as intended, while Direction B's TURN label overrode the same sentence and made readers paraphrase the story the wrong way. A's portrait study failed when cropped into a wide horizontal band, but that was a page problem, not an idea problem. The decision was to take A, rebuild the approach page around the portrait study as its own column, and leave export and accessibility work for a later pass.