How Do You Measure a YouTube Sponsorship?

Decide what you are measuring before you book

A sponsorship bought for awareness and a sponsorship bought for signups are the same video and two different campaigns. The measurement has to be chosen first, because nearly all of it has to exist before the video is published. A promo code cannot be added afterwards, and a branded search baseline cannot be taken retrospectively.

What the creator can see that you cannot

YouTube Analytics belongs to the channel, not to the sponsor. The creator can see watch time, audience retention across the video, traffic sources and viewer demographics. You can see none of it unless they share it.

Ask for it in the agreement rather than after the video is live. The most useful number in the whole campaign is retention across the sponsored segment, because it separates how many people loaded the video from how many were still watching when your product appeared. It is also the number nobody volunteers.

The three things you can measure yourself

• A promo code. The only attribution that needs nothing from the viewer beyond the intent to use it, and the only one that survives a viewer switching device. Give every creator their own code so the credit is unambiguous.
• A tracked link. A UTM-tagged URL in the description, ideally pointing at a landing page built for the campaign. Expect it to under-count. Most viewers never open a description, so a link on its own will make a campaign that worked look like one that did not.
• Branded search. Your own Search Console and Google Trends, before and after. This is the closest thing to a real read on awareness, and the one most brands forget to take a baseline for until the campaign is already live.

Views are the denominator, not the answer

A view count tells you the size of the room. It does not tell you what happened in it. The reason to care about views is that every other number you have is a rate, and a rate needs a denominator you trust.

That is the practical argument for booking against a guaranteed figure rather than an estimate. Every show on our brand partnership roster publishes one. Across 64 shows they run from 10,000 to 2.6 million, with a median of 100,000 and half falling between 56,500 and 150,000. If the denominator is agreed in writing before anything is booked, then a code redemption rate or a click rate means something on the day the campaign ends rather than three months later.

Views keep arriving after you stop looking

A YouTube video is not a flight. It does not stop when the budget does. Search and suggested traffic keep delivering views for months, and an evergreen review or explainer can take a year to finish earning them. Any read you take in the first week is a partial one.

This is also why a guarantee has to say what happens if the target is missed. Ours is agreed up front - either a pro-rated adjustment, or we keep running content until the figure is reached.

It also means the CPM you are quoted and the CPM you end up paying are two different numbers. A campaign priced against a guaranteed 100,000 views that goes on to earn 140,000 has not just delivered more reach, it has delivered it at a lower cost per thousand than the one on the quote. That gap is the whole argument for booking against a guarantee rather than an estimate.

What a real result looks like

Eight recent brand partnerships are published on the roster with the views they delivered, from 57,176 to 416,847, median 162,228. They are listed by brand and by creator so the numbers can be checked against the videos themselves rather than taken on trust.

We publish them for the same reason we publish guaranteed figures. A range you can inspect is more useful for planning than a case study built around the one campaign that went unusually well.

Why we do not publish benchmark conversion rates

You will find click-through and redemption benchmarks quoted confidently across the industry. Most trace back to no stated sample, and the ones that do disagree with each other by several multiples. A number that cannot be sourced is not a benchmark, it is a guess with a decimal point.

What is worth comparing is your own campaign against your own baseline - the same product, the same landing page, the same offer, measured before and after. That comparison is the only one where you control the variables.