Video Production KPIs That Show Where Your Team Loses Time

The monthly report I sent our biggest retainer client in my video-editing agency tracked exactly two video production KPIs, videos shipped and hours logged per editor, right, and both numbers went up every month. What it never showed was how many rounds each video went through, or how long a cut sat with the client before anybody pressed play, so when their marketing head asked me on a call why one short explainer had taken most of a month, all I could say was that I would check with my editor.

That was the mistake, and to be very honest it cost me more than one awkward call. I spent the next week digging through email threads and WhatsApp messages and found that my editor had turned around every cut within a day or two, right, while the video sat in an inbox waiting for three people on the client side to find time to watch it. I had walked into that call half ready to blame the fastest person in the whole chain, and I'm pretty sure that editor felt it, because the trust took a long time to come back.

So this post is basically the five numbers I wish I had tracked from day one, and how I pull each one out of a review tool like PlayPause into a plain sheet, so the next time somebody asks where the month went, you actually have an answer.

Content KPIs versus video production KPIs

Most KPI lists for video are really about content performance, right, views, watch time, click-through rate and so on, and they describe what happened after the video went live. If that is the side you need, there is a separate post on establishing KPIs for video content and another on the three creative marketing metrics leadership tends to ask for.

Video production metrics describe the machine that makes the video, the stretch of post-production work between the first cut and the final file. They answer plainer questions, like how many of this week's days my team spent working versus waiting on somebody else.

The catch here is that production numbers are the ones you can actually move this week, right, because you cannot make an audience watch longer by tomorrow, but you can absolutely cut the days a video sits unopened in a client's inbox, and in my experience that is where most of the lost time hides.

Production KPIs measure the machine

Views and watch time show how a video performed after launch. Production KPIs show how long it took to make and where it sat waiting.

The five production KPIs worth tracking

I tried tracking a dozen things once and it collapsed within a month, because nobody wants to fill in twelve columns every Friday, and these five survived because together they map a video from the first share to delivery.

Videos shipped

This is plain throughput, the count of videos that reached Delivered in a week or a month. It is the number leadership asks for first, which is exactly why I used to report it alone, right, and on its own it tells you volume and nothing about how hard each video was to get out the door.

Revision rounds per video

This is how many versions a video needed before approval, so if MV1 went out and the client approved MV3, that video took two rounds. A rising count is the best early warning I know, because it usually means the brief was thin or the feedback was scattered, long before anybody misses a deadline.

Time in client review

This is the total days a video spent with the client across every round, from the moment a version was shared to the moment notes or an approval came back. In my agency it is almost always the biggest number on the sheet, and it is the one I ignored in that monthly report.

Time to first comment

This is how long after a version is shared the first note lands, and it sounds small, but it separates a client who never opened the link from one who watched on day one and then took a week to agree internally on what they wanted.

Approval to delivery

This is the days between final approval and the delivered file, covering exports, captions, extra aspect ratios and so on, and since it sits entirely on the team side, when it creeps up I know the problem is mine.

1Brief signed and MV1 shared

2Client opens the link and leaves notes

3New version stacked as MV2 or MV3

4Status moves to Approved

5Final file delivered

Measuring revision rounds per video

First settle what counts as a round, or the number means nothing. My rule is that a round is one consolidated set of client notes that leads to a new version, so four comments on Monday and six more on Tuesday against the same cut are still one round, and an editor fixing a typo on their own does not count at all.

In PlayPause the counting is almost free, because new versions stack on the same card as MV1, MV2, MV3 and so on, right, so the version stack is the round counter. I open the card of every approved video, read the highest MV number and subtract one. It is the same idea as version control for code, one history per file instead of a folder of final, final two and final really final exports.

For the average, and this is purely illustrative math, if ten videos were approved this month and their stacks add up to thirty versions, the team averaged two rounds per video. What I really watch is the spread, because an average of two can hide one client sitting at five rounds on every video, and that client is where the real conversation is.

When a video runs long, side-by-side version compare on Agency and up lets me put MV2 next to MV4, and more often than I would like, the rough cut was fine and the brief quietly moved underneath it. The page on how to reduce revision rounds and the post on reducing revision rounds in corporate production go much deeper, and once you have the number, billing for extra revision rounds gets a lot less awkward.

Measuring time spent waiting on clients

This is the approval turnaround metric, the one that would have saved me on that call. You split every round into two clocks, the days the team was working and the days the video sat with the client, right, and the first time I did this properly, the client clock was the bigger one on most of our videos.

In PlayPause, when I share a version, who-watched analytics shows who opened the link, when and from which city, so the gap between sharing and first open is pure waiting. The gap between first open and the first note in the comment thread is thinking time, and with Slack connected every comment also lands in a channel, which makes that first note easy to spot.

Those two gaps tell very different stories. For instance, if the link sat unopened for four days, that is a reminder problem, and a polite nudge on day two usually sorts it out. If the client opened it within an hour and still took six days to comment, that is an alignment problem on their side, too many stakeholders and nobody owning the decision, and reminders will not help. You see what I mean here, right, one number called time in review was hiding two completely different fixes.

Old way

notes over email and WhatsApp with no idea when the client even watched

With PlayPause

one link, who-watched analytics shows the open and comments sit on the exact frame

Once you can see the waiting, the fix mostly belongs in the paperwork, so I now write a review window into every proposal, notes within two working days of the link going out, as laid out in the proposal template with a review process. Buyers can flip it around with the questions to ask a production company, because a vendor who cannot tell you their typical rounds is probably not tracking them.

Review_Cut_v4.mp4In Review

212160p · ProRes

00:34 / 02:18

SR

Sarah 0:34

Frame-accurate note, everyone sees the exact same thing.

Pulling the numbers from your review tool

I do this every Friday by hand, one row per video in a plain sheet, and it takes about twenty minutes, right, because the numbers already live in the tool.

The backbone is custom statuses. I use In edit, With client, Changes requested, Approved and Delivered, and the team treats moving a card as part of the job, same as exporting. Everything else comes from the version stack, who-watched analytics and the comment thread, and I date approval and delivery by hand the moment a card moves, because I would rather log one date than trust my memory on a Friday.

• Count cards that moved to Delivered
• Read the MV number on every approved card
• Note share date and first open date
• Note the first comment date
• Note approval date and delivery date

Custom statuses, version stacks and who-watched analytics are there from Creator, the entry plan, while side-by-side compare and Playbooks sit on Agency and up, so check the pricing page against what you need. If your editors live in Premiere, the Premiere Pro panel brings comments into the timeline, so clicking a note jumps the playhead right to it, and the Premiere Pro productions workflow post shows where that fits. For the bigger picture, see the video production workflow page and the post-production checklist template.

A worked example with one video

Here is one video run through that sheet, and to be clear the dates are illustrative rather than a real client, but the shape is very very typical of my agency. MV1 is shared on Monday the 1st, the first open shows up in who-watched analytics on Thursday the 4th, and the first comment lands on Friday the 5th.

The editor shares MV2 on Tuesday the 9th, the client comments on Wednesday the 10th, MV3 goes out on Thursday the 11th and is approved on Monday the 15th, and after exports and captions the file is delivered on Wednesday the 17th, right, sixteen calendar days from MV1 to delivery.

Read as KPIs, that is two rounds, since the client approved MV3. Time in client review is four days plus one plus four, so nine of the sixteen. Time to first comment was four days on the first round and one on the second, and approval to delivery was two days, which leaves the team holding the video for seven days in total, weekend included.

16 days

MV1 to delivery in the example

9 days

with the client in the example

2 rounds

before approval in the example

On a sheet that only counts videos shipped, this is one delivered video in a normal fortnight, but the full sheet shows the client holding it for more than half its life, so the fix is a nudge on day two and one named approver, and the editor is not the problem at all.

Using KPIs to fix the process instead of blaming editors

The moment a KPI sheet turns into a leaderboard of editors, people start gaming it, right, they sit on cuts to keep their round count low, and the numbers get prettier while the work gets worse. So in my agencies nobody is ranked on these numbers, ever.

Instead I read the numbers by client and by stage, so if one client averages four rounds and everyone else averages two, I look at the brief and feedback on that account before I look at the editor. If the same notes keep coming back, the logo too small or the captions in the wrong font, that goes into a Playbook for that client, the per-client document of creative style and direction, with a checklist the editor ticks before anything is shared. Trust me on any level, a repeated note written down once is worth more than any KPI meeting.

“

A slow round is almost always a process problem wearing an editor's name tag.

The same thinking works in-house, where the waiting usually comes from approvals bouncing between departments, and with brand marketing teams I keep seeing a tiny team clock next to a huge approval clock, where the fix is usually one person who owns the final yes. At the end of the day, the KPIs tell you which conversation to have, and in my experience that conversation is almost never with the editor.

Frequently asked questions

What are the most important video production KPIs?

For me it is five, videos shipped, revision rounds per video, time in client review, time to first comment, and approval to delivery. Throughput alone misled me for a long time, so I always report it next to the other four, right, because together they show how much the team made and where the days actually went.

How many revision rounds per video is normal?

It depends a lot on the type of work, and I would rather you compare against your own history than chase an industry number nobody can source. In my agency I start paying attention when one client or one kind of video keeps sitting well above the rest of the book, because one heavy account can hide behind a healthy looking overall average.

How do I track the approval turnaround metric without a data team?

Start with a sheet, one row per video, and fill it every Friday. The hard part is knowing when the client opened the link and when the first note arrived, and a review tool with who-watched analytics and threaded comments hands you both, so the sheet becomes a weekly summary instead of detective work through old emails, does that make sense, right.

Should editors be judged on these video team performance metrics?

I would say no, and I learned that the hard way. The numbers are really really good at showing process problems like a thin brief or a slow approver, and really bad at judging individuals, because an editor has no control over when a client presses play. Read them by client and by stage, and fix the system before you ever look at a person.

If you want to try this on your own projects, every PlayPause plan comes with a 7-day free trial, so you can set up your statuses, share a couple of versions and see your first real week of numbers. Have a look at PlayPause pricing and pick your plan.

So yeah. That's my way of saying it.