Artwork Requirements & Graphic Design Questions

Artwork requirements & graphic design questions, answered

Almost every print or signage job that goes wrong goes wrong at the artwork stage, and almost always for one of five reasons: no bleed, wrong size, wrong colour mode, not enough resolution, or missing fonts.

This guide explains each of those in plain language, plus the two questions we now get asked every week — how to supply artwork from Canva, and why an AI-generated image can’t just be blown up to signage size. If you take one thing from this page: send us your file before you finalise it, not after. A five-minute check costs nothing; a reprint costs everything.

Supplying artwork

What does “print-ready artwork” actually mean?

Print-ready means a file we can send straight to the press or printer without changing anything. In practice that is: a PDF at the exact finished size, in CMYK, with the correct bleed, with all fonts embedded or converted to outlines, with images at adequate resolution, and with no stray objects, spot colours or transparency issues.

If any one of those is missing, the file needs work before it can print — and that work is either billable design time or a phone call back to you. Neither is a disaster, but both take time you may not have.

What file format should I supply?

A print-ready PDF is the universal answer. If you work in Adobe software, native Illustrator or InDesign files with all links and fonts packaged are also fine. For photographic-only work, a flattened TIFF at final size works. Send the source file as well as the PDF if you have it — it makes small changes far easier.

FormatGood forNotes
PDF (PDF/X-1a or PDF/X-4)EverythingThe preferred supply format. Self-contained and predictable.
AI (Adobe Illustrator)Logos, signage, vector layoutsSupply with fonts outlined or packaged, and links embedded
INDD (Adobe InDesign)Brochures, booklets, multi-pageMust be packaged with links and fonts
EPSLegacy vector artworkStill fine; PDF is generally preferable now
TIFFPhotographic artworkFlattened, CMYK, at final size and resolution
PSD (Photoshop)Image-based artworkWorkable, but text is far better set in vector software
JPG / PNGImages within a layoutNot ideal as a whole-job supply format — no bleed, no vector text
Word / Publisher / PowerPointContent, not artworkSend as content for us to typeset — these cannot be printed from directly

Why can’t you just print my Word document?

Because Word is built for office printers, not commercial presses. It has no bleed, no CMYK colour control, no reliable font embedding, and it reflows text depending on the printer driver and the fonts installed on the machine that opens it — meaning your file can look different on our screen to yours.

We can absolutely work from a Word document — it is a perfectly good way to send us the words. We just re-set it properly in layout software. Same goes for PowerPoint and Publisher: excellent for content, not for supplying final artwork.

What is PDF/X-1a and why do you ask for it?

PDF/X-1a:2001 is a restricted PDF standard designed specifically for print exchange. It forces all fonts to be embedded, all colour to be CMYK or spot, and all transparency to be flattened, and it disallows the things that cause unpredictable output. In short, it removes the ways a PDF can surprise a printer.

In Adobe software it is a preset in the export dialog — choose PDF/X-1a:2001 (or PDF/X-4 if you need live transparency) and set your bleed. In Canva, the equivalent is the “PDF Print” download option.

How do I send you large files?

Email is fine up to about 10 MB. Above that, use a free transfer service such as WeTransfer, or share a Dropbox or Google Drive link. Send the original file — not a version compressed or “reduced for email”, which strips the resolution you need.

Will you check my artwork before printing?

Yes, always — every supplied file is preflighted before it goes near a press. We check size, bleed, colour mode, resolution, fonts and trim safety, and we come back to you if something needs fixing. Send it early and there is time to fix it calmly.

Bleed, trim & safe area

What is bleed and why does my artwork need it?

Bleed is artwork extended past the finished edge of the piece — usually 3 mm on every side — so that when the job is trimmed there is colour to cut into. Guillotines and plotters have a small tolerance, and without bleed a shift of half a millimetre leaves a white sliver down the edge of your card.

Think of it as insurance. Your business card is 90 × 55 mm, but the file should be 96 × 61 mm with the background running all the way to those outer edges. The trim happens somewhere within the bleed zone, and wherever it lands, the result looks intentional.

How much bleed do I need?

3 mm all round for business cards, flyers, brochures and booklets. 5 mm for posters, corflute and rigid panels. 10 mm for banners and pull-ups. 25–50 mm for tensioned fabric and wrapped edges. When in doubt, ask — the allowance is driven by how the piece is finished, not by its size.

ProductBleedKeep content clear of trim by
Business cards3 mm3–4 mm
Flyers, brochures, postcards3 mm4–5 mm
Booklets (saddle stitched)3 mm5 mm, 8–10 mm on thick booklets
Posters3–5 mm10 mm
Corflute signs, A-frames5 mm15 mm
Vinyl banners (hemmed)10 mm50 mm
Pull-up banners10 mm100 mm bottom, 60 mm top
Fabric media walls25–50 mm100 mm

What is the safe area or quiet zone?

The safe area is the margin inside the trim line where all your important content should sit — text, logos, phone numbers, borders. It is the mirror image of bleed: bleed protects you when the trim falls outside the line, safe area protects you when it falls inside.

Two specific traps. A thin border drawn right on the trim line will end up visibly uneven — it will be 1 mm on one side and 3 mm on the other. And text sitting 1 mm from the edge can be clipped. Keep both well inside.

What happens if I supply artwork without bleed?

One of three things. If your design has a plain white border anyway, nothing — it prints fine. If it has colour or images to the edge, we either add bleed by extending the artwork (which we may be able to do cleanly, or may not), or we scale the artwork up slightly to create bleed, which crops a millimetre or two off every edge.

Scaling up is the usual fallback and it is generally invisible — unless your design has content close to the edge, in which case it gets cropped. That is why safe area matters as much as bleed.

Can you add bleed to my file for me?

Often yes, and for straightforward files it takes a minute and we just do it. Where it becomes real work is when the background is a photograph that ends exactly at the edge, when there is a pattern that has to be extended convincingly, or when the file is a flattened JPG with no layers to work with.

If you have the original editable file, sending it along with the PDF makes this trivial. If all you have is a flattened image, expect us to either scale up slightly or quote a small amount of design time.

What are crop marks?

Crop marks (or trim marks) are the short lines at each corner of a print-ready PDF that show where the finished piece will be cut. They sit outside the bleed area and are not part of the printed piece. Registration marks and colour bars serve a similar purpose for press setup.

Include crop marks when exporting a PDF for print. In Canva this is the “Crop marks and bleed” tick box; in Adobe software it is in the Marks and Bleeds panel of the export dialog.

Why does my artwork have to be the correct size?

Because scaling introduces problems. Scaling up reduces effective resolution and thickens hairlines unpredictably. Scaling from one aspect ratio to another either distorts the artwork or crops it. And a file supplied at the wrong size makes it ambiguous which dimension you actually wanted honoured.

Set your document to the finished size before you start designing. If you are told the sign is 2400 × 1200 mm, build the document at 2400 × 1200 mm (or a clean 1:10 scale of it, clearly labelled), not at A4 “to be enlarged”.

Can you just scale my A4 artwork up to A1?

Mathematically yes — A4 and A1 share the same proportions, so it scales cleanly without distortion. Whether it will look good is a separate question that depends on what is in the file. Vector artwork scales perfectly. Photographs and flattened images lose eight times their effective resolution going A4 to A1.

An A4 file with 300 dpi images becomes roughly 106 dpi at A1 — acceptable for a poster read from two metres, not acceptable for an exhibition panel. Text and logos that were vector remain razor sharp at any size, which is the whole argument for building in vector.

Colour: RGB, CMYK & Pantone

What is the difference between RGB and CMYK?

RGB is how screens make colour — red, green and blue light added together, ending at white. CMYK is how printing makes colour — cyan, magenta, yellow and black ink absorbing light on white paper. They are different systems with different ranges, and some colours exist in one and not the other.

Every screen you have ever designed on is RGB. Every commercial press is CMYK. The conversion has to happen somewhere; the question is whether it happens where you can see it and adjust, or silently at the printer where you cannot.

Why do my colours look duller in print than on screen?

Because your screen emits light and paper reflects it, and because the brightest RGB colours — vivid oranges, electric blues, fluorescent greens, saturated purples — sit outside what four process inks can physically mix. The conversion pulls them to the nearest achievable CMYK equivalent, which is always less intense.

The fix is to design in CMYK from the start so you are looking at achievable colour the whole time, rather than falling in love with a colour that cannot be printed. If a specific colour is critical to your brand, use a Pantone spot ink or accept a close CMYK approximation — and see it printed before you commit to a large run.

Should I convert my file to CMYK before sending it?

Yes, if you can — it means you have seen and approved the converted colour rather than leaving it to chance. If your software cannot do it (Canva free plans, most web-based tools), send it anyway and we will convert and flag anything that shifts significantly.

What is a Pantone or spot colour?

A Pantone (PMS) colour is a pre-mixed ink of a specific formula, printed as its own ink rather than built from CMYK. Because it is mixed rather than simulated, it is exactly the same colour every time, on every press, in every country — which is why brand guidelines specify Pantones.

Spot colours are available in offset printing and in vinyl film for signage. In digital printing and large format, Pantones are simulated in CMYK — close, but not identical. Note also that Pantone publishes separate coated (C) and uncoated (U) values for the same colour, because the same ink looks different on different paper.

What is rich black?

Rich black is black built from all four process inks rather than black alone — commonly C40 M30 Y30 K100. Plain K100 across a large area can look thin and greyish; rich black prints deep and dense. Use rich black for large solid areas and big display type only.

Never use rich black for small text, body copy or fine lines. Four inks have to register perfectly to make a clean edge, and the tiniest shift produces coloured fringing on small type. Keep anything under about 20 pt at plain K100.

How do I get consistent colour across print and signage?

Specify your brand colours as Pantone references, not as “the blue in the logo”, and tell whoever is producing each item what the reference is. Then accept that ink on paper, printed vinyl and powder-coated metal will never be pixel-identical — they are different materials reflecting light differently.

Doing your brochure, banner and shopfront through one supplier helps considerably, because the colour decisions are made once and carried across all three rather than being re-guessed by three different production teams.

Resolution & image quality

What does dpi actually mean?

Dpi (dots per inch) and ppi (pixels per inch) describe how densely an image’s pixels are packed into a physical size. Crucially, dpi is meaningless without a size — the same 3000-pixel-wide image is 300 dpi at 254 mm wide and 30 dpi at 2540 mm wide. Nothing about the file changed.

So “is this image 300 dpi?” is not the right question. The right question is “how many pixels does it have, and how big am I printing it?”

What resolution do my images need to be?

300 dpi at final size for anything read close up — business cards, brochures, flyers, booklets. For large format the requirement drops with viewing distance: 150 dpi for exhibition panels, 100 dpi for corflute and pull-ups, 72 dpi for banners, and as low as 25 dpi for billboards.

Read fromExampleMinimum dpi at final size
Under 500 mmBusiness card, brochure, booklet300 dpi
Under 1 mPoster, exhibition panel, menu150–300 dpi
1–3 mPull-up banner, corflute, window graphic100–150 dpi
3–10 mVinyl banner, shopfront fascia72–100 dpi
10 m+Building sign, billboard25–72 dpi

How do I check if my image is high enough resolution?

Find the image’s pixel dimensions (right-click the file and check its properties or info), then divide by the finished print size in inches. Or in metric: pixels ÷ (mm ÷ 25.4) = dpi at that size.

  • Worked example: a 2400 × 1600 pixel photo printed 200 mm wide. 200 ÷ 25.4 = 7.87 inches. 2400 ÷ 7.87 = 305 dpi. Fine for a brochure.
  • Same photo at 1200 mm wide (a corflute sign): 1200 ÷ 25.4 = 47.2 inches. 2400 ÷ 47.2 = 51 dpi. Too soft for close viewing, marginal at distance.
  • Reverse it to find your maximum size: pixels ÷ target dpi × 25.4 = maximum width in mm.
File size in megabytes tells you almost nothing. A heavily compressed 8 MP photo and a lightly compressed one can differ tenfold in file size while having identical pixel dimensions. Always look at pixels.

Can I use images from my website?

Usually not. Website images are deliberately compressed and downsized for fast loading — often 800–1600 pixels wide, which is roughly a 130 mm print at 300 dpi. They also carry JPEG compression artefacts that become visible when enlarged, and they are in RGB.

Go back to the original photos, or to whoever produced the site, and ask for the full-resolution originals. If they no longer exist, budget for a reshoot or a stock image for anything printed large.

Can you make a low-resolution image high-resolution?

We can enlarge it, and modern AI-assisted upscaling does a genuinely good job on photographic subjects. What it cannot do is recover detail that was never captured — it is intelligently guessing at what should be there, which works for textures and skies and fails for anything precise.

Upscaling works reasonably onUpscaling does not fix
Photographic backgrounds, skies, foliageSmall text inside an image
Textures and out-of-focus areasLogos and brand marks
Moderate enlargements (up to about 2×)Fine linework, maps, diagrams
Images with soft, natural edgesProduct labels and packaging detail
Faces at small original scale
Heavy JPEG compression artefacts

For anything with hard edges — a logo, lettering, an icon, a map — the correct fix is not upscaling but redrawing as vector. It takes design time but the result is perfect at any size, forever.

Vector vs raster files

What is a vector file and why do you keep asking for one?

A vector file describes artwork as mathematical shapes — points, curves and fills — rather than a grid of pixels. Because it is instructions rather than pixels, it can be scaled to any size with no loss of quality at all. A vector logo is as crisp on a business card as on a billboard, from the same file.

Raster files (JPG, PNG, TIFF, PSD) are grids of pixels with a fixed amount of information. Enlarge one and you are stretching that fixed information across more space. This is the entire reason signage and print people ask for vector logos: signage is often metres across, and there is no such thing as a raster file big enough for every future application.

What’s the difference between AI, EPS, SVG and PDF?

All four can hold vector artwork. AI is Adobe Illustrator’s native format. EPS is an older interchange standard, still widely used. SVG is the web vector standard. PDF can hold vector, raster or both — which is why a PDF is not automatically a vector file.

FormatVector?Best used for
AIYesMaster artwork, editable source
EPSYesSupplying logos to third parties
SVGYesWeb use, some signage software
PDFCan be eitherUniversal supply format — but check what is inside it
JPGNoPhotographs only
PNGNoScreen graphics, transparency for web
TIFFNoHigh-quality photographic print

A quick test: open the PDF and zoom in to 1000%. If the edges stay razor sharp, it is vector. If they turn into blocks, it is a raster image sitting inside a PDF wrapper.

I only have a JPG of my logo. What now?

Three options. Go back to whoever designed it and ask for the vector master — most designers keep them. Check any old files, brand guideline documents or archived emails, since the EPS or AI is often sitting in a folder somewhere. Or have it redrawn, which is what we do when the original genuinely cannot be found.

Redrawing is not tracing a bad image and hoping. It is rebuilding the logo properly as clean vector shapes with the correct typeface, correct proportions and correct colour breakdowns, so you end up with a master file that will serve you for the next decade across every application.

Can’t you just auto-trace it?

Auto-trace works acceptably on simple, high-contrast shapes and badly on almost everything else. It produces wobbly, over-noded curves, ragged edges where the original was compressed, and lettering that is a rough approximation of the typeface rather than the typeface itself.

Auto-tracing is fine for a rough placeholder. For a logo that will go on a shopfront at 1.5 metres wide, where every imperfection is magnified, it is not. We hand-redraw anything going to signage size.

Why does redrawing cost money?

Because it is skilled work done by a person, not a filter. Redrawing means identifying or matching the typeface, rebuilding each shape with clean curves and correct geometry, matching colours to Pantone references, and delivering a set of master files in the formats you will need — usually one to three hours depending on complexity.

It is a one-time cost that removes the problem permanently. Every future job — signage, print, vehicle, embroidery, web — uses the vector master and needs no further work.

Fonts

What does “outline your fonts” mean?

Outlining (also called converting to curves or converting to paths) turns live editable text into vector shapes. Once outlined, the text is no longer text — it is artwork — so it cannot shift, substitute or reflow no matter what fonts are installed on our machines.

Always keep an un-outlined copy of your file for yourself before you outline, because outlined text cannot be edited or spell-checked afterwards. Outline the copy you send, keep the editable original.

Why did my fonts change when you opened the file?

Because the font was not embedded in the file and is not installed on our system, so the software substituted the nearest available one. This is the single most common cause of a layout arriving looking wrong — line breaks move, spacing changes, and the whole design shifts.

Prevent it by supplying a PDF with fonts embedded, or by outlining the text, or by packaging the job with the font files included. This is exactly why we ask for PDFs rather than native layout files from people who aren’t set up for print.

Can you use my fonts?

If you supply them for the purpose of producing your job, generally yes — this is normal practice and covered by most font licences as production use. Fonts remain licensed software though, and we do not redistribute them or use them on other clients’ work.

If you would rather not supply font files, outline the text before sending. That sidesteps the question entirely and is the cleaner solution for a final file.

Are Canva fonts available to you?

Not necessarily — many are licensed to Canva for use within Canva, not distributed as files you can hand to a printer. This is precisely why a Canva file must be exported as a PDF (which embeds the fonts as outlines or embedded subsets) rather than sent as a Canva link or a design to be edited elsewhere.

Supplying artwork from Canva

Yes, we print Canva files — regularly, and they turn out fine when they are set up correctly.

Canva is a genuinely useful tool and there is no snobbery here. But it was built primarily for screen output, and its defaults reflect that: no bleed unless you switch it on, RGB colour, and a canvas that will happily be the wrong size for the thing you are ordering. Get those three things right and a Canva file prints as well as anything else.

Can I supply artwork from Canva?

Yes. Set the document to the exact finished size before you start designing, turn on print bleed, keep important content well inside the edges, and download as PDF Print with crop marks and bleed enabled. Done that way, a Canva file is a perfectly good print file.

Where Canva struggles is very large signage (canvas size limits), true spot colour matching, and jobs needing precise control over blacks and overprints. For those we will usually rebuild the artwork — but for business cards, flyers, brochures, posters, pull-ups and most everyday print, Canva is fine.

How do I export a print-ready PDF from Canva?

Set the correct size first, switch on print bleed, then download as PDF Print with crop marks and bleed ticked. The full sequence is below.

  1. Set the correct size before you design. Create a custom size document at the exact finished dimensions in millimetres — for an Australian business card that is 90 × 55 mm. Do not design at A4 and plan to scale it, and do not use a US-sized template.
  2. Turn on the bleed guide. Go to File → Settings and tick Show print bleed. A dashed guide appears just outside your canvas — that is the bleed zone.
  3. Extend backgrounds into the bleed. Drag any background colour, image or shape out past the dashed line so it covers the full bleed area. Anything that stops at the canvas edge risks a white sliver.
  4. Pull important content inwards. Keep text, logos and phone numbers at least 4–5 mm inside the canvas edge on small print, and much further on banners. Nothing critical near the trim.
  5. Check your images. Click each image and look for a low-resolution warning. Canva flags images that are too small for the print size — take those warnings seriously, they are the same maths as everything on this page.
  6. Download as PDF Print. Click Share → Download, choose PDF Print as the file type. Do not choose PDF Standard, JPG or PNG.
  7. Tick “Crop marks and bleed”. This is the step most people miss. Without it, the PDF exports with no bleed regardless of whether you turned the guide on.
  8. Choose CMYK if the option appears. Canva’s paid plans offer a CMYK colour profile on PDF Print downloads; free plans export RGB. If you see the option, choose CMYK. If you don’t, send it anyway and we will convert.
  9. Open the PDF and check it. Confirm the size is right, the background runs into the bleed, nothing important is near the edge, and no placeholder text survived. Then send it through.

Why doesn’t Canva add bleed automatically?

Because most Canva designs are for screens — social posts, presentations, documents — where bleed is meaningless. Print is a subset of what Canva does, so the print-specific settings are opt-in rather than default. They are all there; you just have to switch them on.

This is the single most common problem with supplied Canva files: the design is fine, the export just has no bleed. It is a 30-second fix at your end and it saves us guessing at your intent.

Is my Canva file RGB, and does it matter?

By default yes, it is RGB. It matters most for saturated colours — bright oranges, vivid greens, electric blues — which will print noticeably duller than they appeared on screen. It matters less for photographs, neutral tones and text.

Canva’s paid plans include a CMYK option on the PDF Print download; use it if you have it. Either way, be aware that a colour you chose because it looked vibrant on your screen may not survive the conversion, and consider proofing before a large run.

Can Canva match a Pantone colour?

Not as a true spot colour. Canva works in RGB and hex values, and while you can enter a hex code that approximates a Pantone, the file contains no spot colour information. Anything requiring an exact Pantone needs to be built in professional software with the spot colour properly defined.

If your brand has a strict Pantone, ask us for its CMYK and hex equivalents and use those in Canva consistently. It will be close. For a rebrand or a large campaign where the colour must be exact across many items, have the artwork built properly.

Why is my Canva design the wrong size for my banner?

Usually because Canva’s custom canvas has size limits and very large signage exceeds them, so people design at a smaller size intending it to be scaled. That works only if the aspect ratio matches exactly and the content is scalable — and it usually loses resolution.

For large signage, either design at a clean scale (1:10 is standard — a 3000 × 1000 mm banner designed at 300 × 100 mm) and tell us the scale explicitly, or let us build it. Remember that at 1:10 your images need ten times the resolution to survive the enlargement, which Canva’s own warnings will not account for.

Can I use Canva stock images on a large sign?

Often not at full size. Canva’s stock library is sized generously for screen and standard print, but a photo that is comfortable on an A4 flyer may be well short for a 3-metre banner. Canva’s low-resolution warning is your first check; if it flags, believe it.

Also check the licence. Canva’s content licence covers a broad range of commercial use but has restrictions — notably around using elements as a logo or trademark, and around resale of the element itself. If an image is going onto a permanent sign or into a logo, verify the licensing before you commit.

You can, but there are two real problems. Canva’s own licence terms restrict using its stock elements and templates as a trademark, meaning you may not be able to register or exclusively own the result. And Canva outputs raster or limited-vector files, which is not the master artwork a logo needs.

A logo is the one piece of artwork that will be used at every size from a 10 mm embroidered polo badge to a 3 metre shopfront sign, for years. It is the wrong place to save a few hundred dollars. If you already have a Canva logo you are attached to, we can redraw it as proper vector artwork and advise on the licensing.

Should I use Canva or a designer?

Canva for high-volume, low-stakes, fast-turnaround material — social posts, internal notices, simple flyers, event signage you’ll use once. A designer for anything that represents the business long-term: your logo, your brand system, your shopfront, your vehicle, your core brochure.

The honest test is how long the piece will exist and how many people will judge you by it. A one-week promo flyer is a Canva job. A shopfront sign that thousands of people will see every day for the next eight years is not.

AI-generated images

AI image generators are useful. They are excellent for concepts, mood boards, mock-ups and quick visual thinking. Where they cause problems is at the point where somebody wants to put one on a 3-metre sign.

This section explains exactly why that doesn’t work, what “redrawing” means and why it is billable, and where AI images can legitimately be used in print and signage.

Why can’t AI-generated images be blown up to signage size?

Because an AI generator outputs a fixed grid of pixels — commonly 1024 × 1024, sometimes up to around 2048 — and there is no additional detail inside the file to reveal when you enlarge it. Printed 3 metres wide, a 1024-pixel image delivers about 8 dpi. That is a visible grid of 3 mm coloured squares.

The resolution problem is the same one that affects any small raster image. What makes AI images worse is that enlargement also magnifies the specific flaws AI generation produces, which are usually invisible at thumbnail size and glaring at signage size.

What specifically goes wrong with an AI image at large size?

Five things, all of which are invisible on a phone screen and unmissable on a sign: garbled text, unresolved detail in hands and hardware, plastic-looking surface texture, soft or invented edges where a hard edge should be, and repeating artefacts in backgrounds and patterns.

  • Text and lettering. AI generators approximate letterforms rather than setting type. Words come out as convincing-looking nonsense. On a business card this is a smudge; on a sign it is a paragraph of gibberish that everyone reads.
  • Hands, faces, hardware, machinery. Small structural details are guessed at. Fingers merge, tool heads don’t join to handles, hinges go nowhere. At 100 mm nobody looks. At 2 metres, everybody does.
  • Surface texture. AI images frequently have a waxy, over-smoothed, plastic quality with a characteristic sheen. Enlarged, that reads as obviously artificial.
  • Edges. Signage works on hard, clean edges — a logo shape, a keyline, a silhouette. AI images produce soft, slightly noisy edges. There is no crisp vector boundary to isolate, so the graphic never looks properly resolved.
  • Backgrounds and patterns. Repeating elements — bricks, foliage, crowds, fabric — develop distortions and repeats that become obvious across a large area.

Can’t you just upscale the AI image?

Upscaling improves a soft photographic subject and it is genuinely worth doing on some images. It does not fix garbled text, incorrect anatomy, or the absence of a hard edge — upscalers make existing information smoother, they do not correct wrong information. An enlarged mistake is a larger mistake.

Some generators now produce larger native outputs, which helps with resolution but not with the structural artefacts. It is worth sending us the largest version you can generate so we can assess it properly rather than working from a screenshot.

Can AI images be used in print and signage at all?

Yes, in the right role. They work well as concept and mock-up material, as soft background elements on large pieces where nothing is being examined closely, and in small-format print where the required size is within the file’s actual resolution. They do not work as the hero graphic on a large sign.

Works reasonablyNeeds redrawing or replacing
Concept visuals and client mock-upsAny logo or brand mark
Soft, out-of-focus backgrounds on large piecesAnything containing text or lettering
Small-format print within the file’s real resolutionHero images on signage and banners
Textures, abstract patterns, atmospheric fillsVehicle wraps and shopfront graphics
Social media and web useDetailed products, people, machinery at size

What does “redrawing” an AI image mean, and why is it billable?

Redrawing means rebuilding the concept as clean vector artwork — taking the AI image as a reference for the idea and reconstructing it with proper shapes, correct typography, defined colours and hard edges. The result scales to any size and holds up under close inspection.

It is billable because it is design work performed by a person, typically one to several hours depending on complexity. What you get back is a master file that works at every size forever, rather than a raster image that was only ever going to work at one. Simple graphic subjects redraw quickly; detailed photographic scenes are better replaced with a real photograph or a licensed stock image.

Can you convert an AI image to vector?

Auto-tracing an AI image produces poor results, because auto-trace needs clean hard edges and AI images have soft, noisy ones. What comes out is a mass of wobbly shapes that looks worse than the original. Meaningful conversion is a hand-redraw, not a filter.

The exception is a simple, flat, high-contrast AI graphic — a basic icon or silhouette — which can trace acceptably and be cleaned up. Anything photographic or detailed cannot.

Three reasons. It is raster, so it cannot be used at signage scale or across the range of applications a logo needs. It frequently contains near-copies of existing marks, because the model was trained on existing logos. And its intellectual property status is uncertain, which matters if you ever want to enforce your brand.

AI is genuinely useful for exploring logo directions — generate twenty ideas, find the one that resonates, then have it properly designed and built as vector. That workflow uses AI for what it is good at and a designer for what it is not.

Who owns the copyright on an AI-generated image?

It is unsettled, and in Australia copyright generally requires a human author — which puts purely machine-generated output on uncertain ground. Separately, each generator’s own terms of service set what you may do commercially with its output, and those terms differ significantly between platforms.

This is general information, not legal advice. If you are building a brand identity or a long-term campaign around a generated image, check the platform’s commercial terms and get proper advice. The practical risk is not usually that someone sues you — it is that you cannot stop a competitor using something nearly identical.

What should I send you if I made my concept with AI?

Send the AI image itself at the largest resolution you can generate, plus a note about what you want kept — the composition, the colour feel, the subject, the mood. We will tell you honestly what can be used as-is, what needs redrawing, and what would be better solved with a photograph or stock image.

Using AI for the concept stage is completely legitimate and often faster than describing an idea in words. We just need to know that is what it is, so we can plan the production properly instead of discovering it at the printer.

Logos & branding

What logo files should every business have?

At minimum: a vector master (AI or EPS), a print-ready PDF or EPS in CMYK, a transparent PNG for screen use, and single-colour and reversed (white) versions for situations where full colour won’t work. Plus a written record of the colour references and the typeface.

  • Vector master — AI or EPS. The source everything else comes from.
  • CMYK version — for print and signage.
  • RGB / hex version — for web, screen and social.
  • Transparent PNG — for quick screen placement over backgrounds.
  • Single colour (black) and reversed (white) — for embroidery, engraving, one-colour print and dark backgrounds.
  • Colour references — Pantone, CMYK and hex values written down.
  • Typeface names — so future material stays consistent.

If you don’t have this set, it is worth having it assembled once. It removes a recurring cost and delay from every future job with every future supplier.

What’s the difference between a logo, a brand and a style guide?

A logo is a mark. A brand is the whole system — the logo plus the colours, typefaces, imagery style, tone of voice and how they are applied. A style guide is the document that records the rules so anyone producing material for you gets it right without asking.

You don’t need a 60-page brand manual. A two-page guide covering logo usage and clear space, colour references, typefaces and a couple of application examples is enough for most small businesses, and it pays for itself the first time you use a new supplier.

Do I own the artwork you design for me?

Ownership and licence terms should be set out in the quote or agreement for the job, so ask us at quoting stage and we will confirm it in writing. Broadly, a client commissioning original design work expects to own the finished artwork, and that is how we work — but get the specifics in writing rather than assuming.

Worth understanding: licensed stock images and licensed fonts used inside a design remain licensed to the terms of their supplier, not transferred to you. That is normal industry practice and applies whoever does your design work.

Do you keep my artwork on file?

Yes. Job files are archived, so reprints and revisions can be produced from the original artwork rather than starting again. Tell us roughly what the job was and when, and we will find it.

That said, you should hold your own master files too — particularly your logo suite. Any business should be able to walk into any supplier with its own artwork.

Our design service

Do you do graphic design, or only printing?

Both — graphic design has been part of the business since 1989 and most of what we print and manufacture, we also designed. Brand development, logos, stationery, brochures, flyers, campaign material, signage design and vehicle layouts are all in-house.

The practical advantage of design and production under one roof is that the design is made for the process it will be produced by. A brochure designed by someone who has never seen a folder or a guillotine looks fine on screen and disappoints in the hand.

What affects the cost of design work?

How much exists already, how complex the piece is, how many rounds of changes are needed, and whether images have to be sourced, shot or redrawn. A brochure laid out from supplied text and existing brand assets is a very different job to a brand identity created from nothing.

  • Starting point — existing brand and vector logo, or starting from scratch.
  • Content — supplied final copy and images, or content that needs writing and sourcing.
  • Complexity — a single-sided flyer versus a 24-page booklet or a full vehicle wrap layout.
  • Revisions — most quotes include a set number of rounds; heavy redirection after approval adds time.
  • Asset work — logo redraws, image retouching, AI-image reconstruction, stock licensing.

What should I give you to start a design job?

The finished size and quantity, your existing brand files, the final text, any images you want used, examples of what you like, and — most usefully — what the piece is actually meant to achieve. The last one shapes the design more than anything else on the list.

Final text is the biggest time-saver. Design that begins with placeholder copy always gets rebuilt when the real words arrive and turn out to be twice as long.

How many rounds of changes do I get?

Ask when you get your quote — it should be stated there. Most straightforward jobs include a couple of rounds of amendments, which is enough for a normal review process. Substantial changes of direction after approval are quoted separately.

The way to keep rounds down is to gather everyone’s feedback into one consolidated response rather than sending changes in a trickle as each person reads it.

Not sure if your file will print properly?

Send it through before you finalise it. We will check the size, bleed, colour and resolution and tell you exactly what needs fixing — no charge, no obligation, and a great deal cheaper than finding out afterwards.

JPG Design · 1 Glen Kyle Drive, Buderim QLD 4556 · (07) 5456 4499 · Send us your artwork