How we measure this

Most benchmarks in influencer marketing come from one of two places. Either a survey asking marketers what they think they do, or one software company looking at its own customers. We do neither. We pick a category, track every brand in it, and read what the platforms actually show.

This page explains what that means in practice. It covers where the data comes from, how we build a report, what each figure counts, and what our numbers cannot tell you.

surveys used

650+

brands tracked

reports published

strategies detected

Where the data comes from

We use four sources. All of them are public activity on the platforms. None of it is self-reported by a brand.

Creator posts and profiles

Public posts, follower counts, engagement and view counts from Instagram and TikTok.

We collect this straight from the platforms for every creator in a category. We recalculate it each time we refresh, rather than reusing older figures.

Live ad creative

Ads running on Meta, taken from the public Ad Library. This includes the creative itself, when it started, and whether it is still live.

We take every active creative for every brand in the category, not a sample. That means the counts are exact, not estimates.

Creator-tagged posts

Posts where a creator has tagged a brand. This covers both paid and unpaid creator activity around it.

We pull these for each brand across the tracking window. That is how we can count creators and views for a brand instead of estimating them.

Brand profiles

The brands in a category, their follower counts, and how often they post.

We define the category first, then track every brand in it. So when we rank a brand, we rank it against the whole set, not a shortlist.

How a report gets built

1
Define the category
We draw the boundary and publish it. Every brand in the set is named in the report, so you can judge whether our definition matches yours.
2
Collect the whole set
We collect every brand in that category, not a sample. That means their profiles, their live ad creative, and every creator post tagging them inside the window.
3
Recalculate every time
We work out follower counts, engagement and view figures fresh at each collection. We never carry an old number into a new report.
4
Rank against the full set
Every rank is out of the whole category, and we always show the total. That way “12 of 97” cannot be mistaken for “12 of 12”.
5
Detect the plays
We compare what we see against the 25 strategies we track, looking at posting frequency, repeat creators, partnership labels and reused creative. Then we report which plays a brand is running.
6
Write it up
A person writes the interpretation. The figures are measured, but the argument about what they mean is our opinion. We keep those two things clearly separate.

What each figure counts

Most arguments about a number are really arguments about what it means. So here are our definitions.

Creator Tagged Post: A post by a creator that tags the brand, inside the tracking window. It counts whether the post was paid or not. We cannot tell which, and we say so wherever it matters.
Total Creator Views: The view counts on those tagged posts, added up. We record them as the platform reported them on the day we collected. Views keep rising after we look, so treat every view figure as a minimum.
Engagement rate: Interactions divided by followers on a creator’s recent posts. When a platform hides likes, we use views instead and say so on the card. We never mix the two without labelling it.
Active Meta ads: Creatives showing as live in the Meta Ad Library on the day we collected. This is a count of ads, not a spend figure. The Ad Library does not publish spend for most advertisers, and we do not estimate it.
Rank (e.g. “12 of 97”): Position against every brand in the defined category, not against a curated peer set. The denominator is always shown so you can see how big the set is.
Tracking window: The period a report covers. It is stated on the report itself. Windows differ between reports because categories move at different speeds, so do not compare a figure across two reports with different windows.

What our data cannot tell you

Every dataset has limits. Knowing ours is what makes the rest of it useful.

We measure behaviour, not spend

We can see that a brand runs 26 live creatives. We cannot see what it paid for them. If a spend figure appears anywhere in our reports, it came from the client, not from us.

Paid and organic are not always separable

A creator post with no partnership label may still have been paid for. When we say a post is unpaid, we mean it carries no partnership label. That is a statement about the label, not about money.

A snapshot is not a trend

Most figures are one collection at one moment. A brand with zero live ads today may have run a hundred last quarter. Where we make a claim over time we say which two dates it spans.

Categories are judgement calls

Deciding that 97 brands count as protein coffee is a decision we make, and we publish it. A different boundary would change every rank in the report. We always list the full brand set so you can check ours.

Platform data has gaps

Hidden likes, private accounts, deleted posts and regional Ad Library differences all leave gaps. We would rather show a smaller number we can verify than a bigger one we had to guess at.

Strategy detection is inference

We work out which play a brand is running from the pattern it leaves behind, such as how often they post, whether the same creators repeat, and whether creative gets reused. Clear patterns are reliable. A brand doing something quietly at low volume can be missed, so if we do not detect a play, that is not proof it is not happening.

Corrections and questions

If a figure about your brand looks wrong, tell us and we will check it against the collection and correct the report if we got it wrong. If you would rather not appear in a public report at all, tell us and we will remove you. You can reach us at hello@sup.co.

See it applied: the published reports · the 25 strategies we detect

Want this run on your category?

We run the same method on your market. That means every brand in your category, every live creator ad, and the plays running around you.

Get your free report ›