Graphic Designer interview questions

I would define a small set of recognition assets and prove their behavior across the 60 real touchpoints before writing the final guideline.

• The foundation includes brand architecture, logo family, typography roles, color tokens, image behavior, iconography, and a modular layout system.
• I prototype at least packaging, website, social, sales deck, retail signage, email, and one low-cost local production case instead of approving hero mockups alone.
• A rule matrix marks what is fixed globally, locally adaptable, or approval-only across the 9 product lines and 14 markets.
• The handoff includes production masters, templates, asset metadata, accessibility rules, and named owners for exceptions.

I would build responsive logo variants rather than force one detailed master into all four conditions.

• The family includes a primary lockup, compact lockup, symbol, wordmark, and one-color positive and reversed versions with explicit minimum sizes.
• Clear space is tied to a stable feature such as the cap height of one letter, while minimum size is set from physical and pixel tests, not proportions alone.
• The 16 px symbol is redrawn optically, the 24 mm mark gets a production sample, and embroidery is tested at the vendor's stitch count and thread weight.
• Masters are supplied as AI, EPS, SVG, PDF, PNG, and approved color-separated artwork with no user-built lockups.

I would use an endorsed transition unless evidence shows that immediate replacement will preserve recognition in every region.

• Research measures unaided recognition, trust transfer, legal ownership, and customer overlap for the parent and each of the 4 regional brands.
• The transition system defines endorsement lockups, relative scale, shared typography and color, and exact retirement dates by market.
• I test signage, packaging, invoices, app icons, and distributor materials because those assets expose different replacement costs and recognition risks.
• A dated migration kit includes old-to-new mapping, temporary artwork, inventory depletion rules, and a DAM status for every legacy mark.

I would keep the construction grid as a reference but define an approved optical master for actual use.

• I compare the mark at 20 mm, 48 px, and large display size because negative space and perceived weight change with scale.
• Overshoots, curve tension, and visual centering are adjusted in Illustrator, while guides preserve the measurable relationship to the base geometry.
• The guideline shows the approved master, not instructions for users to recenter it, and explains that bounding-box alignment is not the visual alignment point.
• Alignment templates expose named optical anchors for common lockups so every team uses the same correction.

I would set rules per logo variant and validate them against the smallest and most crowded of the 25 applications.

• Clear space uses one internal unit, such as the symbol's counter width, so it scales consistently across horizontal and stacked versions.
• Minimum sizes differ by variant and medium, for example 24 px for the symbol and 120 px for a detailed horizontal lockup after legibility tests.
• Partner lockups specify separator, baseline, maximum partner-mark height, and a protected area that neither brand may enter.
• A contact sheet shows passing and failing examples for app bars, labels, lanyards, signage, and sponsorship walls.

I would separate quick production rules from deeper rationale so 300 users can ship correctly without reading a 150-page PDF.

• A searchable web guideline covers logo, typography, color, imagery, layout, accessibility, rights, and channel-specific examples with downloadable masters beside each rule.
• PowerPoint, Figma, social, document, and email templates lock legal and structural elements while stress-testing long names, missing images, and translated copy.
• A one-page checklist handles routine output, while new patterns and high-risk print, packaging, or public campaigns require named approval.
• Version, owner, effective date, and change log appear on every guideline section and template package.

I would create a controlled premium tier inside the master identity instead of a visually independent sub-brand.

• The master mark, core type family, and signature recognition asset remain fixed, while material, image treatment, spacing, and one accent palette carry the premium signal.
• I test the tier on pricing page, proposal, welcome kit, card, event signage, social, email, and support documents to prevent a hero-only solution.
• Side-by-side tests verify that customers identify both the premium difference and the parent relationship within a few seconds.
• The guideline names prohibited shortcuts such as defaulting every premium application to black and gold.

I would normalize perceived prominence through optical zones and a finite set of approved lockup templates.

• Partner marks fit within square, horizontal, or tall bounding zones with maximum cap-height and width relationships to the host mark.
• Separator, baseline, clear space, background, and monochrome behavior are fixed, while optical adjustments are stored in approved partner assets.
• I build a 40-logo contact sheet at final sponsorship-wall and digital-header sizes to catch marks that look dominant despite equal bounding boxes.
• The DAM distributes prebuilt SVG, PDF, and PNG lockups so local teams never reconstruct them from loose files.

I would create production-specific detail tiers that preserve the crest's recognizable silhouette and key internal cue.

• The master retains full detail, while medium and micro versions remove fine lines, close counters, and tonal modeling at documented thresholds.
• Each process gets a physical or vendor proof using its real line weight, trapping, foil spread, screen mesh, or engraving tolerance.
• The 32 px avatar is pixel-reviewed at 100% rather than judged enlarged, and one-color versions do not rely on grayscale separation.
• Variant names and permitted processes are embedded in filenames and DAM metadata to stop the full crest reaching unsuitable production.

I would deliver one controlled source package plus clearly labeled ready-to-use outputs for print, office, and screen.

• The source contains outlined and live-type Illustrator masters, approved spot and process swatches, construction layers, and a read-only reference PDF.
• Print folders provide PDF and EPS in CMYK and Pantone variants; digital folders provide optimized SVG and PNG sizes in sRGB.
• Filenames encode brand, lockup, orientation, color mode, background, and version, such as Brand_Primary_H_RGB_Dark_v03.svg.
• Checksums, release notes, font licenses, and a DAM canonical link make the source of truth unambiguous to all 26 teams.

I would define shared functional roles while letting each script use a native, optically compatible type family.

• Native readers and type specialists evaluate proportions, texture, weight mapping, numerals, punctuation, and cultural suitability instead of forcing superficial resemblance.
• The specification maps display, heading, body, caption, data, and legal roles with sizes, leading, tracking, and fallback stacks per script.
• I test dense packaging copy, responsive web text, PowerPoint substitution, and 200-page editorial rhythm with real translations.
• Font files, web licenses, embedding rights, supported glyph sets, and fallback behavior are documented before rollout.

I would preserve role relationships while setting medium-specific sizes and measure rather than scaling one layout proportionally.

• Digital type uses responsive tokens with tested minimums, while reports use point sizes and leading appropriate to paper and reading distance.
• Event-wall lettering is calculated from viewing distance and then checked full-size in the venue, not inferred from an A4 proof.
• The same semantic roles, such as display, section, body, and caption, remain recognizable even when absolute sizes differ.
• A specimen sheet shows all roles in mobile, desktop, print, and environmental examples with pass and fail cases.

I would build a modular grid and semantic style system that handles all 6 content types without page-by-page formatting.

• A 12-column grid, baseline grid, and defined margin zones support narrative pages, financial tables, pull quotes, charts, and image-led sections.
• Paragraph, character, table, cell, and object styles encode hierarchy; no-body-copy formatting depends on manual overrides.
• Parent pages and section starts control folios, running heads, language labels, and repeated compliance elements.
• Preflight flags overset text, missing links, RGB images, unresolved style overrides, and minimum type-size violations across all 160 pages.

I would create fixed information zones and flexible modules that preserve variant recognition under the longest bilingual copy.

• The front panel reserves stable areas for brand, product name, variant cue, quantity, and required claims, aligned to dieline-safe zones.
• The back uses a baseline and column grid for ingredients, instructions, nutrition, barcode, recycling, and legal copy with controlled overflow rules.
• I stress-test the smallest pack, longest product name, 35% copy expansion, and darkest variant before templating all 36 SKUs.
• InDesign styles and data-linked fields reduce manual divergence, while each size receives its own approved master rather than scaled artwork.

I would specify repeatable image behavior and metadata so 2,000 assets remain recognizable and searchable without looking identical.

• Photography rules cover subject distance, camera angle, lighting ratio, depth, crop behavior, color grade, and prohibited stock-image cues.
• Illustration and 3D use compatible composition, perspective, palette, texture, and shadow rules while retaining medium-specific craft.
• Test shoots and sample renders prove crops for 1:1, 4:5, 9:16, 16:9, and wide banners before annual production starts.
• DAM metadata records creator, rights, territory, expiry, subjects, campaign, crop availability, and approval status for every asset.

I would define a physical reference and medium-specific targets rather than promise one numeric value will match everywhere.

• The system records a Pantone reference, separate coated and uncoated CMYK builds under named ICC profiles, LAB target and tolerance, and sRGB and HEX values for screens.
• Textile and signage vendors provide material samples because ink, substrate, finish, and illumination change the perceived red.
• Contract proofs and press drawdowns are compared under D50 lighting to the approved physical standard.
• The guideline states which value belongs to each process and forbids converting the HEX value directly into production CMYK.

I would compare the brand value of the orange against the measurable cost and reliability of adding a spot plate.

• A Pantone drawdown and a contract CMYK proof show the actual gamut difference on the chosen stock under D50 light.
• The printer prices 4-color and 5-color runs, including makeready, waste, and any finishing interaction across 50,000 copies.
• If the orange is a primary recognition asset, I recommend the spot ink; if it is secondary, an approved in-gamut substitute may be the better system choice.
• The final swatch, separation rule, and approved proof become production references for repeat runs.

I would separate expressive brand colors from functional text colors and publish only tested pairings.

• Normal text combinations meet at least 4.5:1 contrast, while large text meets 3:1 under the applicable WCAG thresholds.
• Signature colors that fail remain available for fields, illustrations, borders, or large marks, not small text.
• Tokens name purpose such as text-primary, surface-accent, and border-muted rather than exposing all 18 colors as interchangeable choices.
• A Figma matrix and coded reference show approved light and dark pairings, hover states, and pass or fail contrast values.

I would add redundant cues so variant meaning never depends on hue alone.

• Each variant receives a distinct name, short code, pattern or icon, and stable position alongside its color.
• Simulations for common color-vision deficiencies identify confusing pairs, but physical grayscale and low-cost print proofs verify real reproduction.
• Chart styles combine color with line patterns, markers, direct labels, and sufficient adjacent contrast.
• Packaging shelf tests confirm that all 8 variants remain distinguishable at realistic distance and under retail lighting.

I would use one conceptual family with size-specific masters where optical and production needs diverge.

• The 24 px grid fixes keyline, stroke family, corner logic, optical padding, and pixel alignment, while filled and outlined forms are not mixed casually.
• Print and wayfinding variants specify minimum line width, viewing distance, substrate, and reversal behavior rather than scaling the UI path blindly.
• Every icon is tested without a label, with a label, in positive and reversed color, and against culturally sensitive interpretations.
• Illustrator masters feed named SVG, PDF, and PNG exports, with version and deprecated status stored in the DAM.