Creative approval has a shape the rest of monday does not handle well: the thing being approved is a file, the file changes every round, and the person who has to say yes does not work for you.
The version problem
This single habit prevents most of what goes wrong.
The tempting thing is to replace the file in place so the item always shows the current artwork. It keeps the board tidy and it destroys the only thing you will want later.
When a client says I approved that and you say you approved round two, the argument is settled by having round two still in existence with a date on it. If you overwrote it, you have a current file and an approval that appears to cover it, and no way to show otherwise.
Keep every round. Name them so the order is obvious — poster-r1.pdf, poster-r2.pdf — and let the item carry all of them. Disk is cheaper than a reprint.
Rounds
Creative work rarely fails; it goes round again.
Resist a status per round. Round 1 sent, Round 1 back, Round 2 sent… grows without limit and tells you nothing you could not get from a date.
Use the state of the work, plus a number:
Three labels handle any number of rounds, the board stays readable, and which jobs are on round four becomes a filter rather than an archaeology exercise. If a job is on round five, that is a conversation about the brief, and the number is what starts it.
The client
Design around that rather than fighting it.
monday gives you three free guests per account and charges for the fourth. An agency with a dozen clients is buying seats so people can look at a JPEG and say yes.
The seat is not even the real cost. A client who has to create an account, set a password and learn a tool in order to approve a poster will put it off — and your proof sits for four days, which is the actual expense.
Whatever you use, the test is the same: can the person approve it from their phone, in the reply to an email, without an account? If not, expect to chase.
Being honest about it
Sometimes the right answer is not an approval tool at all.
If your reviewers need to mark up the artwork itself — pin a comment to a spot on the page, draw on it, compare two versions side by side, check colour separations — then you want a dedicated proofing product, and no approval app is a substitute. Those exist, they are good, and they cost accordingly.
If what you actually need is the right people said yes, in order, against this version, and I can prove it — that is an approval problem wearing creative clothing, and a proofing suite is a large amount of software to buy for it.
Questions
Send them the approval rather than the file, and make the request carry the version. The failure mode you are guarding against is a forwarded email with an attachment from three rounds ago, and the only reliable defence is that the thing they click is tied to a specific version rather than to a mailbox.
Then the approval no longer covers what is attached, and somebody has to notice. If that somebody is a person, eventually nobody notices — this is the single most common way a wrong revision reaches print with an approval apparently covering it.
A yes is enough for approval. Require words only on changes requested — sending work back with nothing to act on is how a round takes a week instead of an hour.
What we built
Not a proofing suite, and it does not pretend to be.
The artwork is previewed inside the approval request rather than linked. The client gets a signed link — no monday seat, no login, no per-approver cost — sees the proof, and approves, asks for a change, or declines from their phone.
Every decision records the exact file version that was on their screen. And when the artwork is replaced after somebody approved it, the approval reopens rather than quietly covering a proof nobody agreed to, keeping the original decision as a record of what that person actually saw.
No markup tools, no colour separations. If you need those, buy the thing that does those.
Three approvals a month, free, including the external approver. Enough to put a real proof in front of a real client and see what happens.