Track every creator with a unique link, code and QR code, add a post-purchase survey, and calculate ROI = (gross profit from sales minus total cost) ÷ total cost. We recommend measuring at 7, 30 and 90 days, not just the first week: across 10,000+ integrations, about 40% of views arrive 30 days or more after go-live. Compare creators on CPA and effective CPM, keep the winners and book them again.
The metrics that matter
| Metric | What it tells you | How to calculate it |
|---|---|---|
| Views | Real reach of the video | YouTube Studio, from the creator |
| Effective CPM | What you really paid per 1,000 views | Cost ÷ actual views × 1,000 |
| CTR | How many viewers acted on the mention | Clicks ÷ views |
| Conversion rate | How well your landing page sells | Sales ÷ clicks |
| CPA | Cost of each customer | Cost ÷ attributed sales or sign-ups |
| ROAS | Revenue for every unit spent | Attributed revenue ÷ cost |
| ROI | Profit after paying for the campaign | (Gross profit minus cost) ÷ cost |
Views and CTR tell you about the creator and the integration. Conversion rate tells you about your landing page. CPA and ROI tell you whether to do it again. Judge each creator on the metric that matches your goal, and do not mix them: a creator with a low CTR can still have the best CPA if their audience buys more.
How to track every sale
No single method catches everything. Use several and add them up carefully, without counting the same sale twice.
- Unique link with UTM parameters. One per creator, in the first line of the description and in a pinned comment. Your analytics tool then shows visits and sales by creator.
- Short or vanity link. Something like yourbrand.com/creator, easy to say out loud and type. It redirects to the UTM link.
- Unique discount code. It catches people who buy later or on another device. It also gives viewers a reason to act now.
- QR code on screen. Viewers who watch on a TV cannot click. A QR code lets them buy from their phone without typing anything.
- Post-purchase survey. One question at checkout: “Where did you hear about us?”, with the creator’s name as an option. It catches the buyers who used neither link nor code.
- Branded search. Check searches for your brand name during the campaign compared with the weeks before. It is not exact, but a jump is a signal.
Codes and links always undercount. Many viewers buy days later, on another device, or type your name into Google. Your survey answers and branded searches show part of that gap. The exact share varies too much by product to give a reliable figure.
The long tail: why the first week lies
A sponsored YouTube video is not a post that disappears in 24 hours. It stays on the channel and keeps getting views from search and recommendations. Agentio, a YouTube sponsorship platform, analyzed more than 10,000 integrations run on its platform and found that:
Source: Agentio, Mar 6, 2026. First-party data from its own platform; Agentio sells YouTube sponsorships.
The same Agentio analysis found that CTR rises about 10% with each repeat integration with the same creator. That is why we recommend measuring at three moments: a first reading at 7 days to catch problems (broken link, wrong code), a real reading at 30 days, and a final one at 90 days. And it is why links and codes must stay active for months, not just the campaign week.
The ROI formula
ROI = (gross profit from attributed sales minus total cost) ÷ total cost × 100
- Total cost includes the creator fee, usage rights, product samples, production extras and the cost of managing the campaign, in-house or through an agency.
- Gross profit is revenue minus the cost of the product. Using revenue instead makes every campaign look better than it is.
- Attributed sales are the ones you can link to the creator: link, code, QR and survey answers, without duplicates.
If your customers buy again, also calculate it with the customer lifetime value. A subscription app that loses money in the first month can be very profitable by month twelve.
A worked example
Illustrative numbers, not a real campaign. You pay 5,000 for an integration. The video gets 120,000 views in 90 days, so the effective CPM is 41.7. It drives 900 clicks (CTR 0.75%) and 150 attributed sales (conversion rate 16.7%), so the CPA is 33.3. Each sale leaves 60 in gross profit: 9,000 in total. ROI = (9,000 minus 5,000) ÷ 5,000 = 80%. If you had stopped measuring at day 7, with a fraction of those views, the same creator would have looked like a loss.
Want to estimate the cost side before you start? Use our YouTube sponsorship calculator and the market ranges in how much a YouTube sponsorship costs.
Turn results into decisions
- Rank creators by CPA (or by ROI if order values differ a lot).
- Rebook the top third. Repeat integrations tend to perform better as the audience gets used to the recommendation.
- Fix before you drop. A creator with good CTR and poor sales may have a landing page problem, not an audience problem.
- Replace the bottom third with new creators of the same profile as your winners.
This is how a test becomes a channel you can plan. Track it right from the brief: our brief template has a section for links, codes and placement. And if you want a team to run it for you, see how we manage YouTube sponsorships.
Frequently asked questions
How do you calculate the ROI of a YouTube sponsorship?
ROI = (gross profit from attributed sales minus total cost) divided by total cost, times 100. Total cost includes the creator fee, rights, product samples and management. Use gross profit, not revenue, or the ROI will look better than it is.
How do you track sales from a YouTube video?
Combine a unique link with UTM parameters, a unique discount code, a short or vanity link for people who type it, a QR code on screen for TV viewers, and a post-purchase survey asking where the customer heard about you.
When should I measure a YouTube sponsorship?
Take a first reading after 7 days, a real one after 30 days and a final one after 90 days. Agentio found across 10,000+ integrations that about 40% of views arrive 30 or more days after go-live.
What is a good CPA for a YouTube sponsorship?
One below the gross profit per customer, and ideally below the CPA of your other paid channels. There is no universal benchmark: it depends on your price, margin and how long customers stay.
Why do codes undercount YouTube sales?
Many viewers buy days later, on another device, or search the brand directly without using the link or code. A post-purchase survey and a look at branded searches during the campaign fill part of the gap.
