A like costs nothing. Someone taps it while the post is still moving past, and by the evening they could not tell you what it said.
A save is different. A save is a promise to come back, and people only make that promise about something they expect to need again. That is why a carousel is worth building at all: it is the one social format that can hold enough to be worth returning to, and the only one where the reader has to keep choosing to continue.
That choosing is the whole design problem. Every slide is a small decision to swipe or leave, and you get between eight and ten of those decisions in a row.
The cover carries one promise
The cover slide has one job, which is to say what the reader will have after the last slide that they do not have now.
One promise. Not three. A cover that offers “everything about pricing, positioning and packaging” offers nothing, because the reader cannot picture the end state.
Set it in the largest type the layout will take. If you have to shrink the line to make it fit, the line is too long, not the type too big. Cut words until it fits at the size you first chose.
Leave the logo off. Your handle is already above the post on every platform that shows a carousel, and a logo on the cover spends the most valuable space you have on information the reader already has. A single small mark on the last slide does the same work without costing you the opening.
A useful test before you build anything: write the cover line and the recap line on the same page. If the recap does not deliver what the cover offered, you have either the wrong cover or the wrong deck, and it is cheaper to find that out on a blank page than after eight slides of layout.
One idea per middle slide
The middle slides are where good carousels come apart. The usual failure is a slide with two ideas on it, because the writer had ten ideas and eight slides and did the arithmetic.
Do the arithmetic the other way. Count your ideas, and let that decide the slide count.
Each middle slide takes one idea, states it in a heading, and supports it in two or three lines. The heading should make sense read alone, because plenty of readers will only read the headings on the first pass and then swipe back for the bodies.
Keep the layout identical across the middle slides: same heading position, same body position, same margins. An identical layout lets the reader stop parsing the design and start reading the words. When only the words move between slides, the words are what gets noticed.
Then give the deck a running thread. A small counter, a repeated phrase at the same spot, a colour that shifts one step per slide. Something that tells the reader where they are and that there is still somewhere to go. Without it, slide four and slide seven feel the same, and a reader who feels no progress stops swiping.
The second slide deserves its own attention, because it is where most decks lose people. Slide one made a promise and slide two has to start paying it immediately. If slide two is a definition, or an introduction, or a paragraph about why this topic matters, the reader has been asked to wait, and waiting is not what a thumb does. Put the first real idea there and save the context for later, or drop the context entirely, which is usually the right answer.
The close is two slides, not one
Finish with a recap slide: the headings of the middle slides, listed, in order, small. This is the slide people screenshot, and it is the reason the deck gets saved rather than merely finished. You are handing them the whole argument on one screen.
Then the last slide asks for exactly one thing. Follow, or read the linked piece, or reply with your version. One. A final slide with three asks reads as a menu, and people do not order from menus they did not intend to open.
Word budgets, decided in advance
Write to a budget per slide and the design stops fighting you.
A cover holds about eight words. A middle-slide heading holds about six, and its body about thirty. A recap slide holds one line per item. The final ask holds a sentence.
The budgets also stop the deck from growing. Ideas arrive while you are building, and each one looks small enough to squeeze onto an existing slide. A written budget turns that into a visible choice: this idea needs a slide of its own, or it does not belong in this deck. Both answers are fine and the squeeze is not.
These are not laws, they are the numbers that keep type large enough to read on a phone held at arm’s length on a train. Set your own once, write them down beside the template, and then write to them rather than writing first and shrinking the font until it fits. Shrinking the font is how a carousel turns into a screenshot of a document.
Where the deck gets built
In Markowill, a carousel is authored as two to ten slides in swipe order built from one saved canvas template, optionally with a wide photograph fused across the whole deck so slide one shows its left edge and the last slide its right edge. Each deck is produced twice, as a set of slide images and as a PDF, since the image carousel and the LinkedIn document post want different files.
Where to start
Take the carousel you published most recently and open it on a phone. Read only the headings, in order, and then read the last slide.
If the headings do not make an argument on their own, or if the last slide asks for more than one thing, you have found the version worth rebuilding this week.
