SHIPPED — MASTER IN USE
Deckforge
Designing a slide master that carries a brand is straightforward. The one I built shipped clean, looked correct, and was wrong in 48 of its 115 text roles — every box measured against a line height that no real font produces. Nothing appeared broken, which is exactly why it would have survived. What the system is built around now is the measurement that catches that.
48 OF 115 TEXT ROLES WRONG
6 COULD NOT FIT A SINGLE LINE OF THEIR OWN STYLE
MEASURED CORRECTLY
CONTEXT
A template does not hold a brand. PowerPoint hands every user a surface and a set of handles, and the drift starts on the second deck: a headline nudged down two points so a sentence fits, a colour picked out of an older file, a margin set by eye. Nobody breaks the system deliberately. It erodes.
So the slide stops being a surface someone arranges and becomes an artefact the brand definition compiles. One definition holds colour, type scale, grid and lockups. Sixty-three layouts read from it. No layout carries a value of its own, and type sizes are fixed — nothing scales to fit, because auto-fitting text is the reason deck families come apart. Each slide finds its own size, and after three editors the series is gone.
That decision is the whole system, and it is the one that failed quietly.
Line spacing was expressed as a percentage — of the font’s line height, not of the type size, which is what the layouts had assumed. Inter Tight carries 1.21 em, a serif more like 1.30. Every box in the set was therefore around twelve percent too tight for what it was meant to hold, and six roles could not fit a single line of their own style. The text did not clip. It grew into its neighbour.
Fixing the six visible cases would have been the wrong repair. Line spacing is now absolute: type size times factor, resolved before the file exists and independent of which font the recipient has installed. Every box holds its designed line count again, and not one box was enlarged. The price is real and worth stating — absolute spacing ignores font metrics, so a face with unusually tall ascenders will set tight. That is a visible error rather than a counted one, which makes it the better of the two.
MY ROLE
Sole author. The brand definition, both stages of the pipeline, and the measurement underneath them.
Delivered and in use. The artefacts below are rendered from the current master with the client’s names, content and images removed — grid, colour and type scale are untouched.
INTERACTION MODEL
01 AUTHOR
02 THE DEFINITION
03 COMPILER
Owns the definition and the words.
One file is the contract.
Builds the master, then sets content against it.
—
Sets colour, type scale, grid and lockups once, in one file.
—
Chooses the layout whose shape fits the point being made.
—
Writes to the budget of the role, rather than resizing the role.
—
Every layout reads its values from it; none carries its own.
—
Each text role carries a measured character budget, not an estimate.
—
When input doesn’t fit the master, the build stops and says why. It never quietly approximates.
—
Stage one generates 63 layouts from the definition.
—
Stage two places content without touching a single layout.
—
Refuses the run when content exceeds what a role was measured to hold.
THE SPECIMEN
All 63 layouts out of one definition, with the measured type scale beside them.
VIDEO, 24 S LOOP THE FIGMA FILE OF THE MASTER: ALL 63 LAYOUTS, THEN THE CAMERA ZOOMS INTO SINGLE SLIDES (A6, B3i, C4, F6c) AND SHOWS WHAT EACH ONE IS ALLOWED TO DO: FIXED SIZES, NO AUTO-FIT, A CHARACTER BUDGET PER ROLE
THREE LAYOUTS UP CLOSE
Neutral blind content. Every placeholder shows its prompt, so the structure is visible rather than claimed.
A6
Cover, statement
The cover without an image. The project name stands large at the top right, the title sits at the bottom left. The title is fixed at 68 px with auto-fit off; its box holds two lines, about 85 characters, measured against the master’s own typeface. The dashed frames are the placeholders, shown with their prompts.

B1

F6c
THE PIPELINE
One definition, two outputs: the deck and the feed.
The master was built for an aviation company. The same brand definition now feeds their social channel: a local database of the client’s identity and image library, from which the posts are set in Figma every month.
SOURCE
DEFINITION
OUTPUT
Brand guideline and image library
→
One file
→
Decks and posts
The client’s originals stay where they are. Only what a post needs is pulled.
Colour, type, grid, lockups. What isn’t in the guideline isn’t used, even if a template offers it.
The deck master, and monthly posts for LinkedIn with an X adaptation.
The client isn’t named here. Names, content and images are removed; the system underneath is the real one.
DECISIONS
Fix every type size and remove auto-fit entirely.
BECAUSE
Auto-fit lets each slide settle on its own size, and three editors later the family has no scale left. Fixed steps reverse the relationship: the text adapts to the system.
WHAT IT COSTS
A hard character budget per role, and that conversation with everyone who supplies content.
Express line spacing as an absolute value in points instead of a percentage.
BECAUSE
A percentage is a percentage of the font’s line height, which lives in a file the recipient may not have. The layouts were all measured as if it were the type size.
WHAT IT COSTS
The setting no longer adapts to the typeface, so a face with tall ascenders sets tighter than its own metrics would ask for.
Repair the cause rather than the six roles anyone could see.
BECAUSE
The boxes were never too small; the lines were too tall. Enlarging the visible six would have left the other forty-two wrong and hidden the reason.
WHAT IT COSTS
A larger change than the bug report asked for, and every measured budget had to be recomputed.
Keep the generator and the deck that consumes it in separate repositories.
BECAUSE
A master that can be edited while fixing a slide is not a master. The separation makes that impossible rather than discouraged.
WHAT IT COSTS
Two checkouts and a round trip whenever a real deck turns up a real layout problem.
RESULT
SCOPE
BUDGET
63 layouts, one master
A character count per role
Seven families, from the cover to the closing. Every layout opens as a real slide in PowerPoint, every text box keeps its fixed size.
Measured against the real font files, not estimated — and the build refuses content that exceeds it.
REACH ME
Say hi.
Next.js, TypeScript, Tailwind. Statically rendered, no backend. Set in IBM Plex Sans and IBM Plex Mono. Built and audited against WCAG 2.2 AA.
SHIPPED — MASTER IN USE
Deckforge
Designing a slide master that carries a brand is straightforward. The one I built shipped clean, looked correct, and was wrong in 48 of its 115 text roles — every box measured against a line height that no real font produces. Nothing appeared broken, which is exactly why it would have survived. What the system is built around now is the measurement that catches that.
48 OF 115 TEXT ROLES WRONG
6 COULD NOT FIT A SINGLE LINE OF THEIR OWN STYLE
MEASURED CORRECTLY
CONTEXT
A template does not hold a brand. PowerPoint hands every user a surface and a set of handles, and the drift starts on the second deck: a headline nudged down two points so a sentence fits, a colour picked out of an older file, a margin set by eye. Nobody breaks the system deliberately. It erodes.
So the slide stops being a surface someone arranges and becomes an artefact the brand definition compiles. One definition holds colour, type scale, grid and lockups. Sixty-three layouts read from it. No layout carries a value of its own, and type sizes are fixed — nothing scales to fit, because auto-fitting text is the reason deck families come apart. Each slide finds its own size, and after three editors the series is gone.
That decision is the whole system, and it is the one that failed quietly.
Line spacing was expressed as a percentage — of the font’s line height, not of the type size, which is what the layouts had assumed. Inter Tight carries 1.21 em, a serif more like 1.30. Every box in the set was therefore around twelve percent too tight for what it was meant to hold, and six roles could not fit a single line of their own style. The text did not clip. It grew into its neighbour.
Fixing the six visible cases would have been the wrong repair. Line spacing is now absolute: type size times factor, resolved before the file exists and independent of which font the recipient has installed. Every box holds its designed line count again, and not one box was enlarged. The price is real and worth stating — absolute spacing ignores font metrics, so a face with unusually tall ascenders will set tight. That is a visible error rather than a counted one, which makes it the better of the two.
MY ROLE
Sole author. The brand definition, both stages of the pipeline, and the measurement underneath them.
Delivered and in use. The artefacts below are rendered from the current master with the client’s names, content and images removed — grid, colour and type scale are untouched.
INTERACTION MODEL
01 AUTHOR
Owns the definition and the words.
—
Sets colour, type scale, grid and lockups once, in one file.
—
Chooses the layout whose shape fits the point being made.
—
Writes to the budget of the role, rather than resizing the role.
02 THE DEFINITION
One file is the contract.
—
Every layout reads its values from it; none carries its own.
—
Each text role carries a measured character budget, not an estimate.
—
When input doesn’t fit the master, the build stops and says why. It never quietly approximates.
03 COMPILER
Builds the master, then sets content against it.
—
Stage one generates 63 layouts from the definition.
—
Stage two places content without touching a single layout.
—
Refuses the run when content exceeds what a role was measured to hold.
THE SPECIMEN
All 63 layouts out of one definition, with the measured type scale beside them.
63 LAYOUTS CONTENT AND NAMES REMOVED, GEOMETRY AND COLOUR AS SHIPPED
THREE LAYOUTS UP CLOSE
Neutral blind content. Every placeholder shows its prompt, so the structure is visible rather than claimed.
A6
Cover, statement
The cover without an image. The project name stands large at the top right, the title sits at the bottom left. The title is fixed at 68 px with auto-fit off; its box holds two lines, about 85 characters, measured against the master’s own typeface. The dashed frames are the placeholders, shown with their prompts.

B1

F6c
THE PIPELINE
One definition, two outputs: the deck and the feed.
The master was built for an aviation company. The same brand definition now feeds their social channel: a local database of the client’s identity and image library, from which the posts are set in Figma every month.
SOURCE
Brand guideline and image library
The client’s originals stay where they are. Only what a post needs is pulled.
↓
DEFINITION
One file
Colour, type, grid, lockups. What isn’t in the guideline isn’t used, even if a template offers it.
↓
OUTPUT
Decks and posts
The deck master, and monthly posts for LinkedIn with an X adaptation.
The client isn’t named here. Names, content and images are removed; the system underneath is the real one.
DECISIONS
Fix every type size and remove auto-fit entirely.
BECAUSE
Auto-fit lets each slide settle on its own size, and three editors later the family has no scale left. Fixed steps reverse the relationship: the text adapts to the system.
WHAT IT COSTS
A hard character budget per role, and that conversation with everyone who supplies content.
Express line spacing as an absolute value in points instead of a percentage.
BECAUSE
A percentage is a percentage of the font’s line height, which lives in a file the recipient may not have. The layouts were all measured as if it were the type size.
WHAT IT COSTS
The setting no longer adapts to the typeface, so a face with tall ascenders sets tighter than its own metrics would ask for.
Repair the cause rather than the six roles anyone could see.
BECAUSE
The boxes were never too small; the lines were too tall. Enlarging the visible six would have left the other forty-two wrong and hidden the reason.
WHAT IT COSTS
A larger change than the bug report asked for, and every measured budget had to be recomputed.
Keep the generator and the deck that consumes it in separate repositories.
BECAUSE
A master that can be edited while fixing a slide is not a master. The separation makes that impossible rather than discouraged.
WHAT IT COSTS
Two checkouts and a round trip whenever a real deck turns up a real layout problem.
RESULT
SCOPE
63 layouts, one master
Seven families, from the cover to the closing. Every layout opens as a real slide in PowerPoint, every text box keeps its fixed size.
BUDGET
A character count per role
Measured against the real font files, not estimated — and the build refuses content that exceeds it.
REACH ME
Say hi.
Next.js, TypeScript, Tailwind. Statically rendered, no backend. Set in IBM Plex Sans and IBM Plex Mono. Built and audited against WCAG 2.2 AA.
