Ask any tool to lay out four things and you get four rounded rectangles with a one pixel border, a title, and two lines of grey text. It is the default answer. That is exactly the problem: a default is a decision nobody made.
We had a services page built from three of these grids stacked on top of each other. Capabilities as cards. Process as cards. Engagement models as cards. Every section the same shape, so every section carried the same weight, so nothing carried any weight. The page was not wrong. It was just silent.
The card is a container, not an idea
A card says: these items are peers, and each one is a small, self-contained unit. That is genuinely right sometimes. A work grid is cards, because each project really is a peer of the others and the image is doing the talking.
It is wrong the moment the content is not peers. Our four engagement models are not four equal boxes; one of them is where most people land. Our five design levels are a ladder, not a set. Put a ladder in a card grid and you have thrown away the only thing that mattered about it.
What replaced them
Rows. A hairline, a large name on the left, the sentence on the right, and space. The name can be 44px because it is not fighting three siblings for a fixed box. The sentence can be a real sentence because it is not being squeezed into a card's dead zone. And the rhythm of the page changes, because rows breathe differently from grids.
The five levels became a climb: each rung brighter than the one before it, the top one the only thing on the page in green. You cannot do that in a card grid. The card grid would have made Old and S-Tier look like equals.
The test
Before you reach for a card, ask what the container is claiming. If the answer is "these are peers and each is a small unit", use it. If the answer is "I have four things and cards are what you do with four things", you have not designed anything yet. You have reached for the default, which is the one thing your competitors also did.
The rounded rectangle is not a style. It is the absence of one.