Logo Lockups Explained: Primary, Stacked, Icon, and Wordmark
A lockup is a fixed arrangement of the parts of a logo — icon, wordmark, sometimes a tagline — with the spacing and proportions decided once and then locked. Hence the name. Nobody rebuilds it, nobody nudges it, nobody retypes the name in something close to the right font.
The reason lockups exist is that a single arrangement never fits every space. The wide horizontal version that looks confident on a website header becomes a two-millimetre smear across the bottom of a business card. The response is not to squeeze it. The response is a second lockup, built deliberately, that belongs in that space.
Most identities need three or four. Most brand books ship either one, which causes improvisation, or eleven, which causes paralysis.
The standard set
Primary (horizontal). Icon on the left, wordmark on the right, aligned optically. This is the default and gets used in roughly seventy percent of applications. If someone has a space and no instructions, this is what they should reach for.
Stacked (vertical). Icon above, wordmark centred below. For square-ish and narrow spaces: social avatars are the obvious one, but also tote bags, app splash screens, the top of a portrait-format poster. The stacked version usually survives smaller than the horizontal, because the wordmark gets the full width rather than sharing it.
Icon alone. The symbol with no type. Favicons, app icons, social profile pictures, embossing, the repeating element in a pattern. This one comes with a caveat worth being honest about: it only works once the audience already knows the brand. A new company using its icon alone on packaging is asking customers to recognise something they have never seen.
Wordmark alone. The name set in its logotype with no symbol. Useful when the icon appears elsewhere on the same surface (already on the page header, so not needed again in the footer), in very wide short spaces, and in contexts where the icon reads as clutter — a legal document, a letterhead, a sponsor row.
With tagline. Optional, and treat it as a distinct lockup rather than "the logo plus some words". Taglines have their own minimum size, almost always larger than the mark's, and it is the single most common thing missing from brand guidelines. Someone applies the primary logo's 20mm minimum to the tagline lockup and the tagline becomes unreadable. Minimum size has to be tested per lockup, not per brand.
That is the set. Adding a fifth and sixth arrangement has diminishing returns and a real cost: every extra option is another decision you are handing to someone who is not a designer and who will make it at 4:50pm.
Why you cannot just scale the primary
Two separate things break when a lockup is used outside its intended proportions.
The first is legibility. A horizontal lockup scaled to fit a narrow column becomes very small overall, because width is the constraint. The wordmark hits its legibility floor while there is still unused vertical space in the layout. The stacked version uses that space and stays readable at the same overall footprint.
The second is optical balance. When a lockup shrinks, the relationship between icon and type does not scale cleanly in perception, even though it does mathematically. Fine type gets thinner faster than a solid icon shape appears to. A gap that reads as comfortable at 80mm reads as cramped at 20mm. This is why lockups intended for small use are often drawn with slightly heavier type weight and slightly more generous spacing than a pure scale-down would give you.
If you have ever scaled a logo down and felt that something was off without being able to name it, that is usually what you were looking at.
Building one properly
The gap between icon and wordmark should be derived from something in the mark, not chosen by eye each time. The usual candidates: the cap height of the wordmark, the stroke weight of the icon, or the width of a distinctive counter.
Pick one, call it your unit, and express every relationship in it. Gap equals one cap height. Icon height equals two cap heights. Clearspace equals one icon width. Now the whole thing is a system, it can be rebuilt exactly by anyone, and it can be documented in a way that survives being redrawn.
Then override it where your eye disagrees. Optical alignment beats mathematical alignment every time, and the two rarely coincide. A circular icon next to flat-sided letterforms needs to overshoot slightly to appear aligned. A wordmark with a descender in it sits differently against a symmetrical symbol than one without. Measure first so you know what you are deviating from, then deviate on purpose.
This is precisely what a construction grid is documenting. Not decoration for the presentation deck — the record of which relationships were deliberate, so that the next person to touch the file knows what they are allowed to move.
Each lockup needs its own rules
This is the part that gets skipped, and skipping it is what causes the problems that surface two years later.
Each lockup needs:
- Its own clearspace value, expressed in a unit derived from that lockup, not inherited from the primary. A stacked mark's exclusion zone is a different shape from a horizontal one's. The method for deriving clearspace is the same; the number is not.
- Its own minimum size, tested rather than assumed, for print and screen separately.
- A stated use case, in plain language. Not "secondary lockup" — "for square and near-square spaces, including social profiles and app icons".
- Its own set of delivered files, in every colour variant. If you deliver the stacked version in full colour only, someone will need it in white on a dark photo and will build it themselves.
Write those four things per lockup and you have most of a usable logo section already.
The mistakes that keep recurring
Rebuilt on the fly. Someone needs the stacked version, cannot find it, and makes it by dragging the icon above the wordmark and eyeballing the gap. It is close enough that nobody objects and wrong enough to be visibly inconsistent next to the real one. The fix is availability: if the file is in the folder, it gets used.
Elements scaled independently. Icon at 110 percent, wordmark left alone, because someone thought the icon looked small. Lock the group. Say it explicitly in the guidelines.
Too many options. Eleven lockups is not thoroughness, it is an unfinished decision passed downstream. Decide which three or four carry the identity and let the rest go.
No tagline rules. Covered above, and worth repeating because it is so common. If the tagline is part of the identity, it gets its own lockup, its own minimum, and its own tested clearspace.
Documenting the arrangement but not the reasoning. "Gap: 12mm" tells someone what to do at one size and nothing at any other. "Gap equals one cap height of the wordmark" scales, survives redrawing, and can be checked.
Doing this with GridMe
Lockups are a design decision — the software does not know where your mark should sit relative to its type. What GridMe helps with is the layer after that: generating the construction geometry for each arrangement so the relationships are visible and measurable, and building the clearspace and size documentation for each lockup on its own artboard rather than one page that pretends the variants are interchangeable.
Download GridMe if you are documenting more than one lockup per project, which, if you are doing this properly, you are.
