ANDRES / design comparison / 7 October 2026
Five blues. One clear recommendation.
Keep the current navy. It gives text the strongest contrast in the existing palette. Keep protection behind photo text and strengthen it on the mobile hero.
Recommendation: retain the current navy and strengthen the mobile hero overlay.
The current design has one mobile paragraph below the modeled contrast target. A stronger mobile wash corrects it in this study. This is a targeted contrast comparison, not a full WCAG certification.
The current design has one mobile paragraph below the modeled contrast target. A stronger mobile wash corrects it in this study. This is a targeted contrast comparison, not a full WCAG certification.
The simple comparison
All five blues can work. E gives us the largest contrast margin with the current text colors.
See exact ratios and the detailed chart
| Alternative | Blue | Offwhite on blue | Blue on cyan / cyan on blue | Normal-sized text |
|---|---|---|---|---|
| A · Legacy wordmark | #11496C | 9.42:1 | 3.95:1 | Adjust small cyan pairings |
| B · Intermediate 1 | #134162 | 10.54:1 | 4.42:1 | Adjust small cyan pairings |
| C · Intermediate 2 | #163A58 | 11.60:1 | 4.86:1 | Pass |
| D · Intermediate 3 | #18324E | 12.87:1 | 5.39:1 | Pass |
| E · Current navy | #1A2A44 | 14.16:1 | 5.94:1 | Pass |
Normal text needs 4.5:1; large text needs 3:1. All five white/blue pairings exceed the enhanced 7:1 target. A/B can work with darker button text and a revised footer-link color. W3C criteria.
Why dark blue supports trust
Published research supports this direction. In a study of 216 people, darker-blue accommodation offers showed a trust-related pathway to booking intention.
Our recommendation combines that research with the strongest measured text contrast in this palette.
Broeder & Snijder, 2019. The study tested different colors and an accommodation context; it does not prove an exact ANDRES-color or conversion result. Read the peer-reviewed study.
Compare the same homepage
Same type, copy, photographs and layout. Scroll each frame to inspect the full page. Static posters keep imagery constant across alternatives.
Desktop: all 15 combinations. Mobile: five current-gradient palettes, flat/no-overlay endpoints A/E, and the E correction. Uncaptured mobile choices are disabled. Conservative model flags are not website grades.
The mobile correction matters
Current E mobile hero paragraph
4.21:1
Conservative source-photo estimate. Actual letter-position estimate: about 4.24:1. Both are below the 4.5:1 requirement for this 18px text.
E with an 80% mobile hero wash
9.23:1
Same paragraph, photo and navy. Only the mobile hero wash changes. All 22 selected placements meet modeled thresholds in this private prototype.
Keep the desktop gradient and stronger mobile careers wash. Flat 72% protection passes the captured desktop examples, but is not a universal mobile solution. No-overlay image text produces flags with every palette.
What this establishes
- Evidence: E gives the most contrast headroom with the current offwhite/cyan system. C/D also pass tested solid pairings. A/B need different small-text pairings.
- Design judgment: E preserves the current design with fewer changes. A lighter blue remains valid with appropriate text colors and overlays.
- Research context: published trust research informs the dark-blue direction. Our ratios measure contrast, rather than trust or conversion.
Method, sources and limits
- The related Cortex/TRIBE v2 creative report is an illustrative scenario, with no model execution performed to prepare it. Its trust, recall and click themes are design hypotheses; scenario percentages are excluded from these scores. Meta model scope.
- Legacy A comes from #11496C in the full ANDRES master wordmark. Icon-only variant: #104A6D. #26AFFF is a separate accent. B–D interpolate rounded sRGB channels; E is the current #1A2A44 site token.
- Exact solid ratios use specified CSS sRGB and alpha compositing. Photo estimates use served source images, recorded crops/positions, actual overlays, and text size/weight/opacity.
- Conservative rectangle minima may include whitespace. The A/E mobile glyph diagnostic confirms the current E paragraph remains below threshold; the E eyebrow and A careers paragraph pass at actual letters.
- JPEG colors are not authoritative pixel colorimetry. Browser resampling, subtle decorative overlays and physical displays are not calibrated. Near-threshold values need follow-up.
- Static posters and reduced motion hold imagery constant. All moving-video frames, keyboard/assistive technology, focus/hover, reflow and all routes need separate verification. Non-text requirements · Moving content requirements.
- Twenty-two placements are representative roles, not a percentage of conformance. No unfavorable result has been removed. No production colors or CMS content were changed.
Baseline: 6b74e2e13b44786af89b20203009e47584c9aa14. Capture manifest · Exact pairs · Modeled results · Glyph diagnostic