Skip to main content

Measure creator campaigns

Creator campaigns get judged on reach, because reach is the number that's easy to get. Views, likes, follower count, maybe a CPM. So the creator with the biggest audience looks like the best performer, gets re-booked, and the cycle repeats.

The problem is that reach measures how many people saw it, not what happened next. A video with 400k views whose comments are full of "this is an ad" did less for you than one with 60k views whose comments are full of people asking where to buy.

Kinn reads both halves — the performance metrics and the comment conversation — for the same set of videos. That combination is what makes the interesting questions answerable.

Before you start​

Connect the specific campaign videos, not the creators' whole channels:

  • TikTok — one URL per video
  • Instagram — one URL per post or Reel
  • YouTube — individual video URLs, or a playlist if you've collected the campaign into one
  • X — the specific post, or a creator's profile

For each, Kinn pulls performance metrics (views, likes, comments, shares) and the comment threads together — which is what lets you rank by reach and reaction in one question.

tip

Indexing the exact campaign URLs rather than the creators' full channels is what keeps the comparison clean. A creator's unrelated back catalogue will otherwise drown the video you actually paid for.

1. Rank by reach and reaction​

The core question, and the one you can't answer with platform analytics alone:

Rank our campaign videos from the last month by engagement, and separately by comment sentiment. Show both, and flag any creator where the two rankings disagree sharply.

That last clause is the whole point. The creators where reach and sentiment disagree are the ones telling you something:

  • High reach, poor sentiment — you paid for exposure to an audience that didn't like it. Sometimes a mismatch of creator and product; sometimes the ad read itself.
  • Low reach, excellent sentiment — a smaller creator with real credibility. Usually your best cost-per-outcome, and usually under-booked.
  • High on both — re-book, and find out what they did differently.

2. Read what the audience actually said​

Sentiment is a summary. The comments are the detail:

For each campaign video, what are commenters actually saying? Group by theme — interest, skepticism, questions, complaints — and give me the most common question asked across all of them.

That last item is quietly valuable. A question that keeps coming up across different creators' audiences is a gap in your messaging, not in theirs. If four hundred people across six videos ask whether it works on Mac, that belongs in the brief for the next round.

3. Compare the platforms honestly​

The same creator often performs completely differently across formats, and the audiences behave differently too:

Compare comment sentiment on our TikTok campaign videos against the Instagram Reels, and against YouTube. Where did the message land best, and where did it get the most pushback?

TikTok comments skew fast and reactive; YouTube comments skew longer and more considered; Instagram sits in between. Don't compare raw sentiment across platforms without accounting for that — compare each creator against others on the same platform first.

4. Decide the next budget​

The question the whole exercise exists to answer:

We spent $50K across these creators last month and have $15K for a follow-up round. Based on engagement and comment sentiment, which creators should we prioritize, and which should we drop? Chart the top five.

Giving Kinn the actual budget and the decision you're making is what turns a descriptive answer into a recommendation weighted toward your constraint. See Writing good queries for why this matters so much.

5. Track it over the campaign, not just at the end​

A campaign measured only at the end is a post-mortem. Measured weekly, it's steerable:

Every Monday at 9am: summarize how our tracked campaign videos performed over the past week — engagement and comment sentiment by creator, plus any recurring questions or complaints in the comments. Flag anything that changed sharply since last week.

See Scheduled tasks. Catching a poorly received video in week one means you can adjust the brief before the rest of the round goes out.

6. Watch for the second-order effects​

Paid content doesn't stay in its comment section:

Since the creator campaign started, has discussion about us increased in Discord or Reddit? Are people mentioning where they heard about us?

Attribution from conversation is imperfect, but "I saw this in a video" turns up in community posts more often than people expect, and it's the closest thing you'll get to organic lift from a paid push.

Common mistakes​

Ranking by views. The most common and most expensive mistake. Reach without reaction is how you re-book the wrong creator twice.

Ignoring the questions. Complaints get attention; questions get ignored. Recurring questions are the cheapest problem you will ever be handed.

Comparing across platforms without adjusting. A TikTok comment section and a YouTube comment section are different populations with different norms.

Indexing whole channels instead of campaign videos. Your comparison fills up with content you didn't pay for.

Measuring only at the end. By then the budget is spent and the brief is fixed.

Next​