In short
A baseline positions a video against comparable posts from its account. Review the median, sample sizes and dates before interpreting a multiple. A high view total is not necessarily outperformance, and this reference is not a commercial benchmark for your entire niche.
Read the scope
Open Competitors, then an account. Statistics use the collected catalog even when the page shows only one batch of videos. Pagination is not the statistical sample size.
The mean sums available values and divides by the number of measured posts. The median represents the middle of that distribution. One large success can raise the mean much more than the median; read both signals alongside post counts and dates.
Understand the reference
To qualify an outlier, the reference requires at least ten comparable videos from the same account, network and source, with similar age and duration. The median limits the influence of one exceptionally large success.
An account-relative multiple helps read local outperformance. It does not establish sales or advertising profitability.
Example: high totals and outperformance
Fictional example: a video with 24,000 views on an account whose comparable median is 8,000 has a 3× ratio if the required sample is available. Another with 200,000 views on an account whose median is 250,000 remains below its reference level. The largest total is therefore not necessarily the strongest local outperformance.
Use the baseline to identify what is unusual for that account, then watch the video to formulate a creative hypothesis. It does not establish a brand’s sales or an entire product category’s commercial potential.
When the reference is missing
A dash or insufficient history signals unknown data. Check collection before comparing accounts. Counters across networks and measurements from different sources may not be directly comparable.
Do not assume the first page of thirty videos is the whole catalog: statistics may include more collected posts. Conversely, a hundred visible videos do not guarantee a hundred usable references for a particular comparison. Age, duration and source can reduce the genuinely comparable sample.