How to Deal With Difficult Video Clients: Six Types and the Fix for Each
Difficult video clients are usually a process problem wearing a personality. Here are the six types I keep meeting in my own agency, and the exact fix I use for each one.
My worst mistake with a difficult video client was treating him as a personality problem, right. He was the one my whole team called difficult, the guy who went quiet for four days and then dropped eleven notes at eleven at night, and instead of fixing the process I absorbed it, put my best editor on him permanently and let the round count run past six. That is the wrong answer to how to deal with difficult video clients, and it cost me a retainer I genuinely wanted to keep, because by the time we delivered, both sides were quietly sick of each other.
What I worked out afterwards is that he was responding exactly the way my process invited him to, right, because I had given him no deadline, no single place to leave a note and no idea who on his side could say the final yes, so he drip fed and went quiet in turns, and then I called him the problem.
So what follows is the taxonomy I use now, six types that keep showing up across my editing agency and the YouTube channels we run, and the process change that takes most of the pain out of each. Some relationships are genuinely not saveable, and I have written separately about when to fire a repeat offender, but to be very honest, most of the clients I used to file under client from hell turned out to be fixable with a better link and a clearer rule.
How to deal with difficult video clients is mostly a process question
Every behaviour on this list shows up at one of four seams, right, the moment you hand over the cut, the moment they collect their notes, the moment somebody decides, and the moment money moves. I call them seams because the post-production pipeline is basically a chain of handoffs from the edit through sound and color to delivery. Nobody has ever fired me over a cut point or a music choice, the frustration builds at the handoffs, where the work leaves my hands and enters a part of the process I usually never designed.
So when a client starts behaving badly, I now ask which seam I am leaving open before I ask anything about them as a person. A client who sends contradictory notes at midnight was usually never told there was a deadline, and a client who reopens an approved video usually never saw a clear approval moment, a pattern I dug into in the piece on clients who keep tinkering. Does that make sense, right, the behaviour is the symptom and the open seam is the cause, and the seam is the part you actually control.
Notes arrive over WhatsApp, email and phone calls, so nobody knows what the current round contains.
Every note sits on the frame it belongs to, in one link, with a status that says whether the round is open.
The ghost who never replies
The ghost is the most expensive type, because your schedule becomes hostage to a person who has not opened the file. You send MV1 on Tuesday, chase on Thursday, chase again on Monday, and the moment you move another project into that slot, the notes land. I used to assume the ghost was ignoring me, and far more often than I expected, they had simply never pressed play.
The fix has two halves, right, and the first is a deadline that exists in writing before the link goes out. My delivery note now reads, word for word, "MV1 is in the link below, please leave every note on the frame by Thursday at 6 pm, and if nothing has arrived by then I will treat this round as signed off and move on to sound and color." That last clause does the heavy lifting, because it turns silence into a decision, and it has changed more client behaviour in my agency than years of polite chasing.
The second half is knowing whether they actually watched, and this is where who-watched analytics on Creator and up earn their keep, because I can see who opened the link, when, and from which city. A link nobody opened gets a follow-up about the deadline, and a link three people opened twice each with zero comments gets a follow-up about who decides. Those are very very different conversations, and I go into both in the post on client communication that stops the chasing.
If the link was never opened, the gap is a missing deadline. If it was opened three times with no comment, the gap is a missing decision maker.
The drip feeder who sends notes one at a time
The drip feeder is rarely malicious, right, they watch the cut in fragments, on a phone, between meetings, and every time something occurs to them they fire it off. The catch here is that eleven messages over three days add up to eleven micro rounds. As purely illustrative math, if each reopening costs an editor fifteen minutes to reload the sequence and rebuild context, that is close to three hours of overhead on one cut before a single note is fixed, which is where the margin quietly dies.
What works for me is making the round a visible object, so the project sits in a custom status called notes open, the client leaves everything inside the link, and when they tell me they are done I switch it to notes complete. Nothing in the timeline gets touched until that switch happens. It sounds rigid, right, and I was worried clients would hate it, but most of them like it, because for the first time they can see where their own feedback stands.
The other half is asking for one uninterrupted pass before any commenting, which cuts the drip rate a lot and feeds into the round discipline on the page about reducing revision rounds, basically the same idea applied across a whole retainer.
The committee with five opinions
You know the version of this, right, the marketing lead, the founder, the founder's spouse and somebody from sales who joined one call in March, each leaving a note that contradicts the last, and then all of them looking at you to sort it out. To be very honest, this is the type I used to handle worst, because I tried to satisfy everybody and ended up with a cut that satisfied nobody.
The rule I hold now is that a committee can comment as much as it likes, but only one person can approve, and that person is named in writing before the project starts. When two notes fight, I do not adjudicate, I reply inside that thread, address the approver by name, ask them to pick one and then wait. On my side, roles do the matching job, so the Manager on the project is the one person who answers the client while the Editor cuts, and the client hears one voice back.
The waiting matters most, because if I settle the conflict myself I have taken on their internal politics for free, and whichever side loses will blame me. Often the real issue is somebody afraid to make the call, which the post on coaching an indecisive client covers, and the structural version of this fix lives on the client approval workflow page.
A committee can comment as much as it likes, but only one named person is ever allowed to approve.
The vague one and the rewriter
These two look completely different and grow from the same root, a note with no anchor, so the vague one says make it punchier, the rewriter watches MV3 and decides the whole opening should change, and in both cases I am handed a feeling instead of a location. My job is to turn that feeling into a frame, right, and everything after that gets easier.
For the vague one, I force the note onto the picture, with a timestamped comment on the exact frame, a range comment across a section, or a drawing straight on the frame, and suddenly punchier becomes the eight seconds from 0:42 to 0:50 drag, which an editor can act on. It works the way SMPTE timecode works inside the edit, because once both sides point at the same frame, the argument about what we are even discussing ends.
For the rewriter, the fix is a paper stage before I cut anything. On Agency and up there is AI transcription with a clickable transcript, so the client reads the interview, clicks a line to jump to that moment, and comments on whatever they want cut, moved and so on before anybody builds a timeline, which I walk through in the piece on doing a paper edit with a transcript. When a rewrite still arrives late, version stacks keep MV1 through MV4 on one card, and side-by-side version compare, also Agency and up, lets me put MV2 next to MV4 and ask which one they actually meant, roughly what version control does for code. Whatever we settle goes into that client's brand playbook on the same plan, so the next video starts from decisions already made.
- Does the note point at a frame or a range
- Does it say what the note is trying to achieve
- Is it from the named approver or a contributor
- Does it contradict something approved in an earlier version
The late payer
The late payer makes people angriest, and in my experience the cause is usually a vague payment trigger rather than bad faith. If the invoice goes out when the work feels done, then feels done is negotiable, and one more round becomes a free way to delay, so the trigger has to be an event both sides can see on a screen, right.
Approval is that event, and this is the line now sitting in every scope document I send: the invoice is issued when the named approver signs off the final version in the review link, payment is due within seven days, and the full resolution master is delivered once payment clears. In practice the approver writes approved on the final version, I set the status to approved, and that is the trigger, and because the rule was agreed in week one, the awkward conversation mostly disappears. The uglier variant, where payment is held hostage to an extra round, I covered in getting paid when one more round is the excuse.
Link lifetime helps too, gently, because on Creator a share link lasts 30 days before the files expire, on Agency 90 days, on Enterprise links never expire, and on every plan I can password protect or instantly revoke a link. So for instance a client who would happily let a project drift across a quarter has a natural window to finish inside, without me sending a passive aggressive reminder.
When fixing the process is not enough
I'm pretty sure most editors keep a bad client longer than they should, and I have absolutely done it myself, so here is the test I apply after one full project on the new process. If the deadlines are written, the notes are anchored to frames, the approver is named and payment is triggered by an event, and that client still burns roughly double the hours of a comparable project, the problem sits in the fit rather than the seam.
At that point the honest move is a price change or an exit, so I raise the rate to match the real cost, and either they accept and the account becomes worth having, or they leave and I get my capacity back, and at the end of the day both outcomes solve the same problem. What I try never to do is keep the old price and quietly resent them, because that resentment leaks into the work, and the team feels it really really long before the client does.
If most of your income comes from two or three accounts this gets harder, and the PlayPause page for freelance editors shows how a solo editor runs this setup without a producer in the middle. Trust me on any level, one problem client held at the wrong price keeps a genuinely good freelance editor stuck at the same revenue for years.
Frequently asked questions
How do I tell a genuinely difficult client from one I have trained badly?
Run one full project on a tight process, right, with a written notes deadline, every note anchored to a frame, a named approver, and payment triggered by approval. If the friction drops, your seams were the problem all along. If they still miss deadlines and reopen approved cuts with all of that in place, you are looking at a fit problem, and no tool fixes a fit problem.
Should I charge more for a demanding client instead of firing them?
Usually yes, and I prefer it to an exit. Work out the real hours the account consumes across a quarter, including the chasing and the rebuilt context, then price the next engagement at that number rather than the optimistic one you first quoted. Either they pay it and the account becomes worth keeping, or they decline and you recover the capacity, and both outcomes are good.
What if the client refuses a review tool and insists on WhatsApp?
They rarely refuse once they see it, because there is no account to create and nothing to install, they open a link in a browser and comment straight on the frame. I stopped copying notes across by hand, so if a note arrives over chat, I reply asking them to drop it on the frame instead, and after a couple of projects the habit usually sticks.
Does a review tool actually reduce the number of rounds?
Not on its own. What reduces rounds is the rule that one round is one closed batch of notes from one named approver, and the tool makes that rule visible, because a status and a version stack are things both sides can see at a glance. Without the rule you get scattered feedback inside a nicer interface, which honestly helps only a little.
If you want to try this, put one difficult client on a single review link, name the approver, write a notes deadline, and watch what the round count does. Every plan on the PlayPause pricing page starts with a 7-day free trial, so you can run a real project end to end before you commit.
So yeah. That's my way of saying it.
Saumyajit co-founded PlayPause after years watching review and approval quietly eat creative teams' deadlines. He writes about the workflow side of video, feedback, versioning, and getting to a clean sign-off.
Related resources
Keep reading
Bring your team into one review space
Centralize feedback, lock approvals, and deliver faster, start free today.
Sign Up for Free