I once sat through a product update where the presenter's slides had a soft sage-green background with slightly darker green headers — a palette that looked genuinely elegant on his laptop screen when he showed it to me the day before. In the actual room, on a projector that had clearly seen a decade of Tuesday meetings, both greens landed as almost the exact same washed-out gray. He talked for twenty minutes in front of what the back row experienced as a mostly blank screen with faint smudges on it. Nobody said anything. Everyone just quietly checked their phones instead.

That deck wasn't badly designed. It followed the standard advice almost to the letter — a restrained palette, one accent color, plenty of white space. It failed for a reason that "use good contrast" doesn't actually explain: there are two completely different kinds of contrast, and only one of them survives a bad projector.

Color contrast and brightness contrast are not the same thing

Sage-green-on-darker-green is a hue contrast — two different colors, similar lightness. To your eye on a calibrated screen, that difference reads clearly because your monitor reproduces color with a wide, accurate range. A projector that's five years past its recommended lamp life doesn't do that. As bulbs age, their color output drifts and narrows before their overall brightness drops noticeably — reds go muddy, blues shift, saturated colors lose the vividness that made them distinguishable from each other in the first place. Two colors that were clearly different in hue on your laptop can end up nearly identical in the room, even though the projector still looks "fine" in every other respect.

Brightness contrast works differently, and it's the one that keeps working. That's the difference between how light or dark two areas are, independent of what color they happen to be — pure black text against pure white, or a near-white headline against a near-black background. A dying projector can mute color all it wants; it still has to render something noticeably darker or lighter, because that distinction is baked into the luminance signal itself, not the color signal that degrades first. This is the actual secret, and it's less "add contrast" than most advice suggests — it's build your hierarchy on brightness difference and treat color as decoration, not as the thing doing the readability work.

Why a "good enough" projector doesn't fix this

It's tempting to assume this only matters for genuinely broken equipment — the projector that's clearly dim, clearly old, clearly a problem everyone in the office already complains about. In my experience, the more common failure is the projector that looks completely acceptable in the room and still can't hold a hue-based palette. A mid-range conference-room unit rated in the low thousands of ANSI lumens (the industry-standard brightness measurement, distinct from the inflated "lumens" numbers some consumer models advertise) handles black text on white without complaint. Ask it to separate two mid-tone colors of similar lightness under normal office lighting, and it starts to struggle well before anyone in the room would call the unit "bad."

This is the part that complicates the usual "just test your slides" advice: testing on the same laptop screen you designed on doesn't catch this at all, because the laptop is the one device in the chain that isn't going to fail this way. You need to see the deck on the actual projector, or at minimum simulate the failure — judging color decisions from a phone photo taken under bright light exposes weak contrast in a way that staring at your monitor never will, since a photo compresses subtle hue differences the same way a strained projector does.

What the fix actually looks like on a slide

Here's where the advice gets concrete instead of just correct. Dark navy text on a pale blue background is a common "professional" combination — and it's a hue-contrast trap, because navy and pale blue can end up closer in perceived lightness than they look on a monitor, especially once you add a slightly warm-toned bulb into the mix. Swap the background to something close to true white or true black, and keep the text at the opposite extreme, and the same slide becomes almost impossible to make illegible, regardless of what the bulb's color output is doing that day.

The accent color can still carry your brand — that's what it's for. The failure mode isn't "don't use color," it's using color to do a job that brightness needs to be doing instead. A chart where two data series are distinguished only by being different shades of the same blue is a hue-contrast decision hiding inside what looks like a perfectly normal chart; on a struggling projector, both series can render as the same dull smear. Separate them by lightness — one dark, one light, or add a pattern or label — and the chart survives equipment that would otherwise flatten it.

Typography quietly undoes this if you're not paying attention

Once the color decision is right, a lighter-weight font can undo the whole fix. Thin type relies on precise edges to read clearly, and projectors that are struggling with brightness also tend to soften edges slightly, through scaling artifacts or a lens that's not perfectly focused at whatever the room's actual throw distance happens to be. A slide with excellent brightness contrast but a hairline-weight headline can still blur into mush from the back row, for a completely different reason than the color problem — which is the same mechanism that makes font pairing work in the first place: contrast in weight does real structural work, not just stylistic work, and it matters more, not less, when the display itself is unreliable.

The practical version: bold or semibold for anything that has to survive a bad projector, regular weight reserved for body copy the audience can lean forward to read if they need to. It's a small adjustment, and it's the one people skip because it feels like a font preference rather than a legibility decision.

Testing this without access to the actual room

Not everyone gets five minutes with the real hardware before they present, and that's the honest limitation here — there's no substitute that's as reliable as seeing the file on the actual unit. Short of that, a few things get you most of the way: convert a screenshot of your slide to grayscale and check whether your key content is still distinguishable from its background with color removed entirely — if it isn't, you're relying on hue contrast that has nowhere to fall back to. Photographing your screen in bright light does something similar, compressing the color range the way a weak bulb does.

None of this replaces the underlying decision, though. A slide with one clear headline and generous space around it reads fast because of hierarchy, not because there's less on it — and brightness contrast is what makes that hierarchy legible in the first place, on a good screen or a bad one. Strip the color out mentally before you ever get near the projector, and the projector stops being able to surprise you.

Next deck you build, try the grayscale test on your title slide before you touch a single font or accent color. If the hierarchy survives with the color turned off, the room's projector — however old, however tired — isn't going to be what decides whether the back row can read it.