The first packaging report I ever sent was two paragraphs long and said the new thumbnail "performed really well." The client wrote back with one question I could not answer: compared to what? That was the day I learned that a packaging report is not a victory lap. It is a comparison, and if you cannot name the thing you are comparing against, you do not have a result. You have a feeling.
Most freelancers and studios get this wrong in the same direction. They report the number that looks best. Click through rate went from 4.1 percent to 6.8 percent, screenshot attached, invoice attached. Nobody mentions that the video also got a fraction of the impressions, which is exactly how CTR goes up while the channel goes nowhere. Clients are not stupid. They notice when the numbers you show and the numbers in their bank account point in different directions, and that is the moment you stop being a partner and start being a vendor.
So here is how I actually report packaging results, what I refuse to claim, and how to write the thing so a client can read it in ninety seconds.
What actually counts as a packaging result?
A packaging result is a change in how many people chose to watch, measured against a baseline you agreed on before you started. Everything else is decoration.
Packaging is the thumbnail and the title and the first frame promise designed together. So the result has to be measured on the unit that the viewer actually experiences, which is the browse or suggested card, not the thumbnail in isolation. A thumbnail that lifts clicks while the title over promises will show a good CTR and a collapsing retention curve, and if you only report the first half of that you are hiding the part that matters.
Four numbers carry almost all of the signal:
- Impressions. Volume of chances. If this moves ten times, CTR comparisons get shaky.
- Impressions click through rate. The choice rate. Your primary metric.
- Views from browse and suggested. Where packaging lives. Search traffic behaves differently and should be split out.
- Average view duration for the first thirty seconds. The honesty check on your promise.
I report all four every time, even when one of them is ugly. Especially when one of them is ugly.
Why does click through rate lie so often?
CTR lies because it is a ratio, and ratios move when either half moves. When YouTube pushes a video into a broader, colder audience, impressions climb and CTR falls even though the packaging is doing more work than before. When distribution narrows to the subscribers who already love the channel, CTR climbs on a thin base.
The practical consequence: never report CTR as a single number without the impression count beside it. I write it as one line. "7.2 percent on 214,000 impressions, up from 5.4 percent on 31,000." That sentence is honest. "CTR up 33 percent" is not, because it hides the fact that the two numbers came from different worlds.
The second trap is time. A video's CTR on day one is mostly a subscriber number. By day fourteen it is mostly a browse and suggested number, and it is usually lower. If you screenshot the good day and call it a result, the client will open Studio three weeks later and see something else. I have a whole piece on how long to run a thumbnail test before the number means anything, and the short version is that you need a stable impression window, not a nice looking Tuesday.
A packaging report that never contains a loss is not a report. It is marketing, and the client can tell.
How do you build a baseline the client will accept?
You build it before you touch a single file, and you get it agreed in writing.
My method is boring and it works. I take the last eight to twelve uploads on the channel, exclude anything that obviously broke the pattern such as a collaboration or a video that got picked up externally, and pull the median CTR and median impressions for the same time window I will use later. Median, not average, because one runaway video will drag an average somewhere useless.
Then I write one sentence that both of us sign off on. Something like: "Baseline is 4.6 percent median CTR across the last ten uploads at fourteen days. We will compare the next eight videos at fourteen days on the same traffic sources."
That sentence does three things. It fixes the metric, it fixes the window, and it fixes the comparison set. Ninety percent of arguments about packaging results are actually arguments about one of those three, and they only happen when nobody wrote them down.
One thing I refuse to promise up front is a target number. I will not tell a channel we will get them to eight percent, because I do not control their niche, their upload cadence, or how the algorithm feels that month. What I will promise is a process, a defined comparison, and unlimited revisions until the packaging is right. Packaging is the highest leverage lever on a channel and it is still not a guarantee machine.
What goes into the one page report?
One page, five blocks, no jargon. If it takes more than ninety seconds to read, the client will skim it and remember only the biggest font.
- The headline sentence. What happened, in plain language, with both halves of the ratio. No adjectives.
- The four numbers, before and after. A small table beats a chart here. Charts invite squinting.
- The packaging decision. Why this thumbnail and this title. Face or object, two words or four, what the first frame promised. This is the part clients actually enjoy, because it teaches them something.
- What did not work. At least one line, every time.
- The next move. One recommendation, not five.
Block three is where you earn the retainer. Anyone can paste a screenshot. Explaining that you dropped the subject to sixty percent frame height because the card renders at roughly 360 pixels wide on mobile is the thing that makes a client trust the next decision before they see the result. If your client has never thought about render sizes, the YouTube thumbnail size guide is worth sending along with the report, and it quietly explains why designing at full size is a mistake that shows up in the numbers weeks later.
How do you report a packaging loss?
You report it first, in the headline sentence, before anything good. Then you say what you think caused it, and what you will change.
A loss usually has one of four causes, and naming which one you believe it was matters more than apologising. Either the promise in the title outran the video and retention punished it, or the thumbnail lost legibility at small size, or the topic itself had no demand and no packaging was going to fix that, or the video simply did not get enough impressions to say anything at all. That last one is the most common and the least discussed. Twelve thousand impressions is not a result. It is noise wearing a percentage sign.
When I am unsure whether legibility was the problem, I go back and run the squint test on both versions side by side and put both crops in the report. A client who can see the failure will forgive it. A client who is told "the algorithm was tough this week" will start shopping around.
Which channels should not be judged on CTR at all?
Small channels and search led channels. On a channel under a few thousand subscribers, impressions are so low and so subscriber weighted that CTR swings five points on nothing, and reporting it as a trend is close to fiction.
For those channels I report absolute views from browse and suggested, plus how many videos crossed a threshold the client cares about. That is a real number that does not evaporate when the denominator moves. I wrote separately about what to measure when you are still packaging for your first 100 subscribers, and the short answer is consistency of visual identity, not percentages.
Search led and archival channels need a different frame again. A documentary style channel may see a repackaged three year old video quietly compound for months, which is invisible in a fourteen day window and obvious in a ninety day one. Match the reporting window to how the traffic actually behaves.
Should the report include old videos?
Yes, and this is usually the most profitable section in it. Repackaging a back catalogue is the cheapest growth most channels have, because the videos already exist, already have watch history, and already have proof they can hold attention.
Old videos also make cleaner evidence than new ones. A video that has been sitting at a stable impression rate for six months gives you a genuine before and after, with the content held constant. New uploads never give you that, because the topic, the length and the upload day all changed too. When a client wants to know whether packaging works on their channel, I would rather repackage four old videos and show them the step change than argue about attribution on a fresh upload for a month.
If a client wants to verify any of this themselves rather than take my word for it, YouTube's own analytics documentation explains how impressions and CTR are counted, including which surfaces are excluded. I link it in reports on purpose. Pointing a client at the primary source is a strange flex that builds a surprising amount of trust, and the official creator updates are worth watching when reporting definitions change under you.
What we do at ClickTheory
We are a YouTube packaging studio, not a template shop. Every thumbnail is custom for that video, styles get tested on our own channels before they go near a client, and title consultation is part of the work rather than an upsell, because you cannot report on a thumbnail whose title is fighting it.
The core package is $250 for 8 custom thumbnails, about $31 each, one per video, with a dedicated designer, unlimited revisions and a 48 to 72 hour turnaround per thumbnail. If we want a clean read on a specific video, an A/B variant is $15, a title pack is $25, and rush is $20. Client names never appear in anything we publish, so results get described by niche and subscriber band only, for example a finance channel around 80k subs. That rule applies to your channel too.
We are Dubai based and work async worldwide, which in practice means reports land overnight for most of the Americas and Europe. If you want to see the work before the spreadsheets, read the pricing and services page.
If you have a packaging report in front of you right now and you cannot tell whether it is good news, send it to samy@pixelsamy.com and I will tell you what it actually says.