How to track views in a clipping campaign

Short answer

To track views in a clipping campaign, read each clip's view count from the public post several times during its tracking window, store every reading instead of overwriting the last one, and pay from the difference between readings. Do not rely on numbers or screenshots sent in by clippers, and do not rely on one count taken at the end.

A spreadsheet works for a handful of clips. Past that, the reading schedule, deleted posts and fraud checks take more time than reviewing the clips themselves. The numbers below use one example campaign: $1 per 1,000 views, a $2 minimum, a $500 cap and a 21-day window. Example settings, not a market average.

What to record for every clip

Clip tracking is a log, not a single number. Each clip needs a fixed record plus a list of readings that only grows:

Platform

TikTok, Instagram Reels or YouTube Shorts. Rates, minimums and caps are often set per platform, so the platform decides which rules apply.

Post URL

The public link to the post, not a link to the clipper's profile. One URL per clip, checked for duplicates.

Verified account

The account that posted it, confirmed as belonging to the clipper who submitted it. Without this, anyone can submit someone else's viral post.

Submission time

The moment the clip entered the campaign. The tracking window and the rules version both hang off this timestamp.

Readings

One row per check: time of the check, views, likes and comments. Never edit an old row; add a new one.

Deleted-post flag

Set when a check finds the post gone, with the time it was first missing.

Review status

Pending, approved or rejected, with who decided, when and why.

Why screenshots and self-reported numbers fail

• The person reporting is the person being paid. A screenshot shows a number on a screen at one moment. It does not show when it was taken or whether the image was edited.
• A screenshot is one reading. It cannot show the pace of growth, which is where botted views show up.
• It does not prove ownership. A screenshot of a popular post says nothing about whether the submitter posted it.
• It does not scale. Every screenshot is a message to open, a number to type into a sheet and a file to keep. Reviewing the clip itself is extra.

Decision rule: accept a URL from the clipper, and nothing else. Every number used for payment is read from the public post by you or by your tooling.

Why one reading at the end of the campaign fails

Reading every clip once, on the last day, looks efficient. It breaks in three ways:

1. Views keep growing. A clip submitted late in the campaign is read before its window has run, and a clip submitted early is read long after its window ended. Either the clipper is underpaid or you pay for views outside the window.
2. Posts get deleted. If a post is gone on the last day, a single reading finds nothing. You cannot pay for what the clip did, and you cannot enforce a rule that posts must stay live.
3. Bought views can arrive late. A final count cannot tell organic growth from a block of views added the day before the check. A reading history shows when each part of the count appeared.

Worked example: one clip, six readings

One example clip, read six times during its 21-day window. Each stage pays the settled total at that reading minus the settled total at the previous one. The minimum (2,000 views) and the cap (500,000 views) apply to the running total, never to a single increment.

Four things the table shows. The first reading pays nothing because the clip is under the minimum. The stages add up to $500, the same as settling 820,000 views once at the cap. The clip stops earning between day 7 and day 14 when it passes 500,000 views; the views after that are reach you do not pay for. With a 10% fee deducted from the clipper's side, the clipper receives $450.

Now suppose the clipper deletes the post after day 14. With only an end reading, you have no count at all. With the history above, you know the clip had 640,000 views on day 14 and that $500 was already settled through that reading. What deletion does to that money is a rule for your brief; the point is that you have the data to apply it.

How often to read clip views

There is no official schedule, and platforms do not publish view curves for short clips. Decision rules that hold up:

• Read at submission. The first reading is the baseline. Views gained before submission are visible in it.
• Read most often while the clip is growing fast. Fraud checks need the pace of growth, and a spike between two readings taken days apart looks the same as steady growth.
• Keep reading after growth stalls. Less often, but until the window ends, so a late spike and a deletion are both caught.
• Take a final reading at the window's end. It closes the clip; views after it are not paid.
• Log failed readings. A private post, a rate limit or a removed video looks the same as zero growth unless the failure is recorded.

Spreadsheet and screenshots vs automated tracking

Decision rule: once you cannot read every live clip at least once a day by hand, move to automated tracking. The time goes into reading and typing numbers, not into judging clips, which is the part only you can do.

The tracking window

The tracking window is how many days after submission views count toward payment. It runs per clip. A clip submitted on day 14 of the campaign is tracked until day 35 with a 21-day window.

A longer window pays each clip for more of its views and spends the budget on fewer clips. A shorter window leaves late views unpaid. Decision rule: publish the window in the brief, start it at submission, and settle each clip when its own window ends, not when the campaign ends. Setting the window is covered in step 5 of how to run a clipping campaign.

Engagement vs views as a fraud signal

Bought views raise the view counter. Unless likes and comments are bought in the same proportion, the ratio between them drops. That makes the engagement rate per reading the first thing to check. A clip with 100,000 views and a dozen reactions is visible the moment likes and comments are stored next to views.

• Compare within the campaign. Same platform, same source material. A clip far below its peers deserves a look before payout.
• Look at the step, not the total. A jump in views between two readings with no matching change in likes or comments is the pattern to check.
• Hold, then decide. Hold the payout and ask the clipper for analytics before you reject.

Buying fake views is not only a campaign problem. The FTC's Consumer Reviews and Testimonials Rule, in effect since October 21, 2024, prohibits buying or selling fake indicators of social media influence, such as views generated by bots. YouTube excludes views from automated click or scroll bots from Shorts revenue sharing, and Whop excludes bot, script and automated views from Content Rewards payouts. More detail: fake views in clipping.

How insy tracks clips

insy is infrastructure for running and settling clipping campaigns, not an agency, and view tracking is the center of it. The brand keeps the brief, the source material and the approve or reject decision; the tracking works like this:

• A separate tracker. Views are read by a specialized tracking system that runs independently of the rest of the platform. Clippers do not report numbers and do not send screenshots.
• Every reading stored. Each check is its own record: views, likes, comments, estimated payout, a suspicious-traffic score and the time of the check. A clip's full history is kept.
• Tracked from submission. A clip enters tracking right after submission. Approval gates the payout, not the measurement, and the campaign balance is recalculated after every reading and every status change.
• Logged cycles and failures. Every measurement cycle is logged, so you can see which clips were not read and why.
• Deleted-post detection. A post removed from the platform is flagged separately, which makes a “keep the post live” rule enforceable.
• Growth signals. Each clip is labeled new, growing, spike, declining or stalled from the pace of its view growth.
• Account verification. Before an account can submit clips, the clipper confirms it with a code, so each submission is tied to a verified account, not just a link.

Payouts are then settled from the difference between two readings, as in the worked example above. The full breakdown is on the clip tracking page; how those readings turn into money is in how to pay clippers. Clippers find campaigns in insy's public campaign catalog.

Common questions