3 days free on every plan · Try any plan before you pay.

Start free trialStart

BlogUGC distribution

How to track views across hundreds of posts in one campaign

Tracking views on an influencer or UGC campaign with hundreds of posts: what to record, when to read counts, which APIs give them, and why numbers differ.

By Abondy ResearchPublished 9 min read

In short: Give every post its own row: the link, the app, the account, when it went up, its latest view count and when that count was read. Read the counts on a schedule instead of collecting screenshots, and judge posts at 7 days. Timing matters by app. In our data, TikTok posts had 82.6% of their later views by 24 hours and Facebook posts 95.2%, but Instagram Reels only 56.2%, reaching 94.8% by day 7. Where you can, pull counts from the apps' own APIs: YouTube's are readable through its Data API, while TikTok's and Instagram's need the creator's permission. Then decide up front how to treat deleted posts and apps that under-count.

A campaign with five creators fits in a notebook. One with hundreds of posts doesn't. Of the 26 campaigns with counted posts made through Abondy from August 29 to November 12, 2026, 10 had at least 100 posts, 5 had at least 1,000, and the largest had 21,920. This guide is what we learned counting them. We make Abondy, whose dashboard does this for posts made through it; most of the method works with any tool or a spreadsheet.

Why is tracking views across many posts hard?

Four things go wrong at scale:

  • Counts keep moving. A number read on day one is not the number on day seven, and the gap differs by app.
  • Each app counts a view differently. YouTube counts a view "the moment a video starts to play" (YouTube Help). Instagram's API counts the "Total number of times IG Media has been played on Instagram" (Meta).
  • Posts disappear. Creators delete posts or set them to private, and the count you saw last week can't be checked again.
  • Screenshots don't add up. They're taken at different ages, they can't be re-read, and nobody can check them later.

What should you record for each post?

One row per post on one app. If a video goes on TikTok and Shorts, that's two rows.

FieldWhy
Post linkThe only way to re-read the count or check the post is real
AppCounts aren't comparable across apps
Account and its follower countTo compare like with like, and to spot one account carrying the campaign
Video or versionTo see which video or hook got the views
Posted atEverything about timing is measured from here
Latest viewsThe number you report
Read atSo you know how old the post was when you read it, and which posts never gave you a number (a blank is not a 0)
StatusLive, private or deleted

Report the median and the mean together. In view data a few posts get most of the views: in our sample of 43,451 posts, the top 1% held 37.9% of all views. The mean (461 views a post) tells you what the campaign delivered in total. The median (35) tells you what a typical post got. Also report the share of posts past a mark you care about, such as 1,000 views.

When should you read the view count?

Read each post more than once, and report the reading taken at 7 days or later. Here's how much of a post's views had come in by each age, in our data.

AppShare of later views counted by 24 hoursBy 48 hoursBy 7 daysMedian post at 24 hours
TikTok82.6%89.8%92.6%95.2%
Instagram Reels56.2%73.4%94.8%60.7%
Facebook95.2%97.5%97.8%98.6%

This is a random sample of 4,000 posts made in October 2026 with at least 100 views, read again when they were at least 7 days old; 2,280 of them are on these three apps. "Later views" means the count at that last reading. The first three columns add up all views; the last is the middle post. Our reading history for YouTube Shorts in this window is too sparse to time this way, so Shorts are left out. For more detail on timing, see when views stop coming in.

What that suggests as a schedule:

  1. 24 hours: a first check that the post is up, public and counting. Don't judge Reels yet.
  2. 48 hours: most TikTok and Facebook posts are close to done.
  3. 7 days: the reading you report.
  4. Later: a last pass for big posts, since a post that took off can keep adding views after a week.

Instagram's own API adds a delay of its own: Meta says "Data used to calculate metrics can be delayed up to 48 hours."

Where do the view numbers come from on each app?

You can read each post's public page by hand, or use each app's API. The APIs differ in who has to say yes.

AppOfficial way to read viewsWho must allow itNotes from the docs
YouTubeData API, videos.list with statisticsNo one else: we read other people's public Shorts this way with an API key, 50 videos a callviewCount is "The number of times the video has been viewed"; a list read "usually costs 1 unit" of a default "10,000 units per day"
TikTokDisplay API, video.list scopeThe creator grants your app the scopeThe scope reads "a user's public videos"; view_count is "Number of views of the video"
InstagramMedia insights, views metricThe creator's professional account, with the insights permission"Data used to calculate metrics can be delayed up to 48 hours"

Sources: YouTube Data API quota, YouTube video resource, TikTok Display API, TikTok video list, TikTok video object, Instagram media insights.

So for a campaign on other people's accounts, YouTube is the easy one. For TikTok and Instagram you either ask every creator to connect their account to your tool, or you read the public counts yourself.

Why don't the numbers match between dashboards?

Most mismatches come from one of these:

  • Different definitions. YouTube's viewCount "counts views the moment a video begins to play" and, per its docs, "includes co-views when multiple people watch together on a television screen." Other apps count plays their own way.
  • Different read times. A creator's screenshot from day two and your reading from day seven are different numbers, both true.
  • Reading the page vs the API. A reader that uses the public page can miss views. Ours can on Instagram, so we treat our Instagram numbers as a floor.
  • Deleted posts. Decide whether a deleted post's views still count. Ours do on the brand side: a campaign keeps the views a post had while it was up, and the post shows as taken down.
  • Bought views. A sudden jump with no likes is worth a look. Our guide to checking whether views are real covers what to look for.

Can you do this in a spreadsheet?

Yes, up to a few hundred posts, if you're strict. Keep the columns in the table above, add a column for the 7-day reading, and fill it only once the post is 7 days old. Then summarise by app: posts, median views, mean views and share past 1,000. Past a few hundred posts, the reading itself becomes the job, which is when an API or a tool that reads counts for you pays off. If the posts promote an app, add installs to the same sheet, as in tracking app installs from TikTok.

What does Abondy's dashboard show?

For posts made through Abondy, the reading is done for you. Here's what the brand dashboard shows, as built.

Each campaign's page:

  • Total views, with the views added this week next to it.
  • A chart with four ranges: 48H (views per hour), 7D and 30D (views per day), and All (a running total since the first post).
  • Posts, credits spent and the budget.
  • The top five posts, each with its app, country, how long ago it went up, its views and a link to the post itself.
  • "See all" opens every post, newest first, 50 at a time. Each row shows the app, the views, the country, when it went up, a link to the post while it's up and one of four states: Checking, Live, Taken down or Credits back.

The Campaigns page has a chart of views for every day since your first post, which you can scroll back to any date.

Your credits history lists each post that didn't reach 100 views 24 hours after its link was handed in, in words such as "Under 100 views on TikTok in 24 hours · refunded", because those posts aren't charged.

The API. GET /campaigns/:id returns the campaign's views, likes, uploads, uploads still up, uploads taken down, posters and a split by app. GET /campaigns/:id/uploads returns one row per post (id, platform, country, posted_at, views, state, url), 50 a page with a cursor and a total, so you can pull every post into your own sheet or BI tool. Webhooks tell you when an upload goes live or is taken down. Calls are limited to 60 a minute. The developers page has the details, and posting to several apps by API shows a full order.

Pricing, for context: a post is 12 credits, 16¢ to 20¢ on a plan from $20 a month; with an API key it's $0.30 a post and $0.05 a video made. For what our posts get month by month, see the views report.

How we counted

  • Posts made through Abondy from August 29 to November 12, 2026 (Eastern time), live or verified when we pulled them on November 21. Flagged, rejected and deleted posts are left out.
  • The sample is mostly gaming clips, largely Roblox, on small accounts. A few campaigns dominate it: one brand's campaign accounts for half the posts, and Abondy's own campaigns for about a quarter.
  • Each view count is the count when we last read it; a few posts were last read before they were 7 days old.
  • 4,210 posts (9.7%) were checked but never returned a view count to our reader, and they count as 0 views in the median and mean above. That share swung from week to week (on TikTok, from none to more than half of a week's posts), which is why a tracker should keep "never read" apart from 0.
  • Instagram counts are a floor, because our reader can under-count them.
  • The timing table uses the 4,000-post October sample described above. Shares by 24 hours, 48 hours and 7 days use the highest reading taken by that age.

Questions people ask

How do you track views from an influencer campaign?

Keep one row per post with its link, app, account, posting time, latest views and the time of that reading. Read counts on a schedule, report the reading at 7 days, and give the median and the mean. Use APIs where creators allow it.

How often should you check views on campaign posts?

At 24 hours to confirm each post is up, at 48 hours for TikTok and Facebook, and at 7 days for the number you report. In our data Instagram Reels had only 56.2% of their later views by 24 hours, so an early Reels number undercounts.

Can you get a creator's TikTok views without their login?

Through TikTok's official API, the creator has to grant your app access to their videos. Without that, you can read the public count on each post's page yourself.

Why are Instagram views lower in my tracker than in the app?

Meta says its insights data "can be delayed up to 48 hours", and readers that use the public page can under-count. Ours can, which is why we treat Instagram numbers as a floor.

Should views from deleted posts still count?

Decide before the campaign starts and write it down. On Abondy's brand side they do: a campaign keeps the views a post had while it was up, and marks the post as taken down.

Read next