When researchers at the US Food and Drug Administration went through seven years of allergen recalls, they found that labelling-associated errors caused 71.1% of the recalls where a root cause was known. That’s 914 out of 1,286, across fiscal years 2013 to 2019. Not every one of those was a proofing failure. Some were the right label on the wrong pack. But plenty start the same way: a line of copy changed late, somebody signed off a version, and afterwards nobody could say who had actually read the ingredients panel.
That’s the part most buying guides skip. They compare viewers and comment tools, and those matter. The harder question, the one that decides whether artwork proofing software earns its keep, is what an approval actually means once five or six people have touched the file.
Start With Who Signs, Not With the Viewer
Before you sit through a single demo, write down everyone who currently has a say in a piece of artwork. For a packaged product the list is usually longer than people expect. There’s the designer and a brand manager, often someone in marketing, a regulatory or legal reviewer for the copy, sometimes an outside agency, and finally the printer.
Now mark what each of them is actually responsible for. The brand manager cares about colour and logo placement. Legal cares about claims and mandatory wording, allergen statements included. The agency mostly wants to know whether its layout survived everyone else’s edits. These are different jobs. They’re often done on the same PDF by people who never speak to each other.
My view is that this map matters more than any feature list. A team that has one can judge a tool in an afternoon. A team without one tends to buy whichever product has the nicest interface, then argue about ownership inside it for the next two years.
Why One “Approved” Button Is Not Enough
Most artwork proofing software records a single state per file: approved or not. That works for a two-person job. It falls apart once approvals start meaning different things.
Picture the brand manager clicking approve on version four. Did they approve the whole file, or only the parts they’re qualified to judge? If legal approved version three and marketing then changed a claim on version four, is the file still legally approved? In a lot of systems the screen says yes, simply because the last person clicked the button.
US regulators have already worked through this for electronic records. Under 21 CFR 11.50, a signed electronic record in FDA-regulated work must show the signer’s printed name and the date and time of signing. It must also show the meaning of the signature, for example whether it was a review, an approval, or a statement of responsibility or authorship. You may not be bound by that rule, and most brands outside pharma and medical devices aren’t. The principle is still worth borrowing. A sign-off with no stated meaning is only half a record.
So when you evaluate tools, ask whether an approval can be scoped. Can a reviewer approve copy only, or brand only? Does a change after legal sign-off reset legal’s approval, or does it quietly carry over? If the vendor can’t answer clearly, the audit trail you’re sold will be accurate and still misleading.
What to Test in Annotation and Version Comparison
Reviewers spend most of their time in annotation, so test it with your real reviewers rather than your design team. Designers are comfortable in almost anything. A legal reviewer who opens proofs twice a month is not.
Four tests worth running during a trial:
- Upload two versions of a real packaging file with one small copy change buried in the back panel. Does the comparison view make it obvious, or do you end up hunting?
- Ask a non-designer to comment on a specific line of text, then upload a new version and check whether the comment still points at the right spot.
- Have two reviewers leave contradictory comments on the same element and see how the tool shows the conflict, if it shows it at all.
- Time how long a first-time reviewer takes to find the approve button without help.
On-artwork markup, with pins and drawn boxes, is nice to have. Honestly, plain text comments tied to a specific version are enough for plenty of teams. What you can’t do without is a comparison view that catches changes nobody announced. Unannounced edits are exactly how an approved claim turns into an unapproved one.
Routing: Sequential, Parallel or Somewhere in Between
Routing decides who sees the file and in what order. Sequential routing, brand first and legal last, is slow, but each reviewer sees a settled version. Parallel routing is quicker and tends to produce conflicting comments that someone then reconciles by hand.
The best setups I’ve seen mix the two. Creative reviewers work in parallel, and the regulatory reviewer goes last, on a version nobody is allowed to change afterwards. If the tool can’t enforce that final lock, you’re relying on people remembering not to edit. That’s how late changes slip through.
Check who breaks a deadlock, too. When brand wants the claim bigger and legal wants it reworded, does the software name an owner, or does the file just sit in review for a week? Some tools let you mark approvers as required or optional. Useful, provided you’ve already decided who is required.
The Approval Record Is What You’re Really Buying
Six months after launch, nobody cares about the proof viewer. What matters is whether you can show who approved which version and for what purpose, and prove the file sent to print was the one they signed.
Ask each vendor to show you an exported approval log for a finished job. You want named individuals rather than shared team logins, with a timestamp on every decision. You also want the final version locked so it can’t be edited after sign-off. Ask how long records are kept and whether you can take them with you if you leave.
Identity deserves a harder look than it usually gets. Many tools send reviewers a link that needs no login, which is convenient and perfectly fine for a café menu. For regulated copy it’s weaker, since anyone holding a forwarded link can click approve. Where the stakes are high, give up a little convenience for named accounts.
Some print platforms now build this record into the job itself. PrintXpand’s artwork approval software, for instance, logs each comment and sign-off against a specific proof version with timestamps and IP details, then locks the approved file before it moves to production. That tie between approval and output is what you want, whichever product you choose.
Where Software Won’t Help
Artwork proofing software can make a clear process faster. It can’t make an unclear one clear. If nobody has decided who owns final sign-off on packaging copy, routing rules just automate the argument, and you get a beautifully documented record of confusion.
It doesn’t replace the printer’s checks either. Once the artwork is approved on the brand side, the printer still has to confirm the file will reproduce properly on press. That’s a separate conversation with its own proof.
So who gains the most? Teams with more than three reviewers per file, and anyone who has already had to explain a reprint they couldn’t trace. A small business sending a flyer to one printer probably needs little more than a shared folder and a clear email.
The pattern we run into most at PrintXpand isn’t a bad viewer or a missing feature. It’s an approval that meant one thing to the person who clicked it and something quite different to everyone who relied on it later. Settle that first and the software choice gets much easier.
Author bio
Pratik Shah is Creative head at PrintXpand, a cloud and on-premise print platform used by 350+ print businesses in 40+ countries. He works with print businesses and brand teams on how artwork moves from design through review into production, with a particular interest in multi-stakeholder approval and keeping a clean record of who signed off on what. More at printxpand.com.
