Media group: how we counted, and how to set it up
For the media analyst who wants to know how visits were credited to a video campaign without tracking anyone, and what the checks were.
- Every definition behind the numbers on the case study
- The setup in the store, step by step
- What to check in your own data
Open the method and setup notes
Enter your work email and the notes open on this page. We also send you the link, so you can come back to it.
Thanks. The notes are open below.
We use your email to know who reads our research. Privacy policy
What was measured
Period and population
A Malaysian media group with more than 20 million users. Two boosts of the same video to a rewards page on Facebook and Instagram in June 2026: 5 to 19 June bought for engagement (19,281 impressions) and 23 to 30 June bought for clicks (46,246 impressions), 65,527 paid impressions in all. Visits were measured to 23 July. Another advertiser's paid click ad to the same page, 10 to 19 July, was read the same way.
What counts as a result
Visits to the destination page above its normal level, credited to each campaign from its own schedule, while it ran and in the weeks after. Clicked visits are not included: about 720 people clicked an ad; about 5,300 watched and came later on their own.
How an order is credited
No orders and no people. A visit is credited by timing: the two weeks before any ad ran set the normal traffic to the page, and visits above that level follow the shape of each campaign's delivery. The method follows measurement methods Google and Meta have published.
Controlled or observed
Observed against a quiet baseline, with three checks on every report: the curve's fit to the ad schedule (a correlation of 0.91), the model finding the other advertiser's end date from the traffic alone, and the peak day at 3.7 times a normal day.
What the numbers do not say
They do not say who came. Nothing in how the campaigns were bought changed; the delivery schedule was read-only. Cost per visit is spread over every campaign-driven visit, not just the clicked eighth.
The terms on the case study
| Term | How it is defined here |
|---|---|
| Baseline | The destination page's normal daily traffic, taken from the two weeks before any ad ran. |
| Credited visit | A visit above the baseline assigned to a campaign from its delivery dates and impressions, during the campaign and after it. |
| Click report | The platform's count of visitors who tapped an ad: about 1 in 8 of the visitors the campaigns brought. |
| After the ads stopped | Credited visits dated after 30 June: about 3,700, seven in ten of what the videos brought. Half the effect was still working ten days after exposure. |
| Visitors per 1,000 impressions | Credited visits divided by paid impressions, times 1,000: about 150 bought for engagement, about 69 bought for clicks. |
| Correlation 0.91 | How closely the credited visits followed the ad schedule, from 0 (no relation) to 1 (exact). |
What was switched on, in order
A quiet baseline
The two weeks before any ad ran set normal traffic to the destination page.
The delivery schedule
Dates, impressions and spend per campaign from Meta, read-only. Nothing changed in how campaigns were bought.
A curve for each campaign
Visits above normal credited to each campaign from its own schedule, including the weeks after it ended.
The checks
Fit to the schedule, the end date found from traffic alone, and the peak day against a normal day.
The read-out
Credited visits per campaign, the share after the ads stopped, visitors per 1,000 impressions by objective, and cost per visit.
To check it in your own store
Take the daily visits to a campaign's destination page for the two weeks before it, the campaign dates, and the daily impressions. Visits above the baseline that follow the delivery shape are the campaign's, clicked or not.
Compare that count with the click report, and check that the visits tail off after the ads stop rather than falling to zero. The gap between the two counts is what a boosted video brings that a click report cannot see.
Back to the result, or see it in your store
See which of last week's visitors you missed
Thirty minutes, with your store and ad accounts open. Then 30 days free. A holdout group decides: if the agent doesn't add orders in 30 days, you don't pay.
Book 30 minutes
You'll talk to one of us.