Measurement · Views, reach and impressions
Views vs reach vs impressions in influencer marketing
Views count viewing events under a platform's rules. Reach describes the distinct audience reported for a defined scope. Impressions count qualifying displays. They describe different parts of exposure, so keep each name attached to its number.
Explore the guideViews
The recorded plays or views
Reach
The estimated distinct people or accounts
Impressions
The recorded displays
Start with this
20,000 views does not necessarily mean 20,000 different people.
Adding reach across posts can count the same audience more than once.
Use the platform's metric definition, reporting period and distribution scope before comparing results.
What each metric tells you
Start with the exact field in the report. Platforms and content formats may apply different counting rules, and some audience figures are estimates. A short-video view, a long-video view and a thumbnail impression should not be treated as identical units of attention.
| Metric | What it describes | What it does not prove |
|---|---|---|
| Views | Qualifying viewing events under the platform's definition | How many different people watched, or whether they watched to the end |
| Reach / accounts reached | The distinct audience counted or estimated within the report's scope | A new audience for the brand, or unduplicated people across platforms |
| Impressions | Qualifying displays of the content or a defined surface | That someone watched, noticed the product or remembered it |
| Unique viewers | A platform's estimate or count of distinct viewers for the selected scope | An exact identity list or a total you can add across separate reports |
Why views and people are different counts
A person may encounter the same content more than once, and repeat viewing may contribute to the view count under the platform's rules. Accounts can also watch several creators in the same campaign. A report can therefore show a large number of viewing events without reaching an equally large number of different people.
Do not rename views as people reached to make a report easier to read. If a distinct-audience measure is unavailable, say so and report the view count you do have. Follower totals cannot fill that gap: some followers may never see the post, while people who do not follow the creator may see it.
Illustrative example · one post, one reporting period
18,000 views and 10,000 reported unique viewers
These figures describe 18,000 recorded viewing events and an estimated distinct audience of 10,000. If both metrics use compatible counting rules and the same scope, the ratio is 1.8 views per reported unique viewer. It does not mean every viewer watched exactly 1.8 times or that the campaign reached 18,000 people.
Preserve the platform's meaning
YouTube's thumbnail impressions count eligible displays of a video thumbnail on specified YouTube surfaces. They are not a count of every way a viewer could encounter a video. YouTube also reports estimated unique viewers separately from views. A thumbnail impression and a video view answer different questions.
When creators send reports from Instagram, TikTok or another platform, keep the original field names rather than translating every exposure figure into impressions. Ask which format and date range the report covers, and whether a screenshot shows a single post, the whole account or a campaign. If a platform changes its counting method, note the change before comparing earlier and later results.
Do not add post reach and call it unique campaign reach
Two posts can reach some of the same accounts. Adding their individual reach counts includes that overlap twice. The same problem applies across creators and platforms, where you may have no reliable way to identify shared viewers.
Use a combined, deduplicated figure only when the reporting method actually supports it. Otherwise label a sum as summed post reach, with audience overlap not removed. Do not estimate the overlap from a feeling that two creators have similar followers.
Illustrative example · overlap assumed known for the calculation
Two posts, some of the same viewers
The simple sum is 18,000, but only 15,000 different accounts are represented in this hypothetical example. In a real report without the overlap figure, you cannot claim that 15,000 total. Report the separate counts and the limitation instead.
| Quantity | Count |
|---|---|
| Distinct accounts reached by post A | 10,000 |
| Distinct accounts reached by post B | 8,000 |
| Accounts reached by both posts | 3,000 |
| Combined distinct accounts | 10,000 + 8,000 − 3,000 = 15,000 |
When average frequency is useful
Average impression frequency is impressions divided by reach when both come from the same compatible reporting scope. It describes repeated exposure on average. It does not show how evenly those impressions were distributed or whether any particular person noticed the message.
Do not divide paid-ad impressions by organic reach, or one week's impressions by a month's reach. Keep views per unique viewer separate from impression frequency: similar arithmetic does not make the underlying events the same.
Illustrative frequency calculation
50,000 impressions across 20,000 reached accounts
50,000 ÷ 20,000 = 2.5 impressions per reached account on average. Some accounts could have seen the content once and others several times. This number alone cannot establish whether the repetition was useful or excessive.
Choose the measure that fits the creator's job
For sponsored distribution, views and reported reach can help describe exposure, alongside viewing quality and audience relevance. For content-only UGC, the first deliverable is the agreed asset: there may be no creator-post reach to report. If the brand later publishes or promotes that asset, measure that distribution separately.
For an affiliate campaign, views can explain part of the journey but cannot establish attributed sales. Use the brand's actual transaction and attribution records for sales conclusions. Do not infer purchases from follower counts, comments or a predicted conversion rate.
- Collect the post link, platform, format and publication date.
- Record the metric label, reporting period and capture date.
- Separate paid and organic distribution where the report permits; say when it does not.
- Keep unavailable reach visible instead of substituting views.
- Use watch time, retention or completion information where available to understand the viewing experience, with their definitions recorded too.
How to describe exposure without overstating it
Write a result that another person can interpret: total recorded post views, the posts included, the date captured and any unavailable audience data. Put mixed platforms or formats in separate rows before offering a campaign total, and explain that a combined view count contains different counting rules when it does.
A first campaign can still be informative without a perfect unique-audience total. State what you observed, what remains unknown and what you will collect next time. The useful improvement may be agreeing a consistent creator-insights handover before the next campaign starts.