A competitor publishes something that does ten times their usual numbers. By the end of the week someone has forwarded it to you with "should we be doing this?"
The instinct is to answer fast. The right answer takes about an hour and usually isn't "yes".
First: confirm it's actually a breakout
Big-looking numbers are frequently an artefact of the comparison.
Compare to their distribution, not yours. A competitor's 200,000-view video is unremarkable if their median is 150,000. It's extraordinary if their median is 8,000. The unit is unusual for them, which is the same outlier logic you'd apply to your own catalogue.
Check the age. Older content has had longer to accumulate. Compare within similar age bands.
Check for paid. A spike on an otherwise ordinary piece — especially one with a hard call to action, or numbers that jumped well after publication — is often distribution they bought. Public data can't confirm it, so hold it as uncertain rather than treating a media buy as a content lesson.
Check the composition. Views without proportionate engagement usually means it was pushed rather than passed along. High engagement per view alongside the views is what makes it interesting.
A surprising share of forwarded breakouts don't survive this step.
The five questions
If it's real, work through these in order. The order matters — each one can end the enquiry.
1 · What actually got shared?
Watch it properly, twice. Then answer: what is the single thing someone would say when passing this on?
If you can't compress it to one sentence, you haven't found the mechanism, and anything you build from it will be a copy of the surface.
Usually it's one of: a claim people found surprising, access nobody else had, a person being unusually candid, genuine production ambition, a useful thing given away free, or a fight someone picked.
2 · Is the mechanism transferable?
Now test that sentence against your own situation.
Some mechanisms transfer cleanly — a format, a structure, a topic that turned out to have demand. Some don't transfer at all, because they depend on something you can't replicate: a founder with an existing following, access to a customer who'd never appear on your channel, a moment in the news cycle, or a controversy you'd be mad to invite.
3 · Was it the content or the moment?
Check the publish date against what was happening in your category that week. A funding announcement, a competitor's failure, a regulatory change, a news story.
Timing-dependent breakouts are real and worth noting, but the lesson is "be ready to respond to moments", not "make more videos like this one." Those imply completely different plans.
4 · Is it one, or is it a pattern?
Look at their previous five and next five pieces.
If it's isolated: more likely a fluke or a moment. Wait. If they've found something, they'll do it again within a quarter — and their second attempt tells you far more than their first.
If the pieces after it changed: they think they found something and are reallocating. That's a competitive fact worth acting on, and it's visible from the format mix shift rather than from the breakout itself.
If it's their third in a row: it's a strategy, and you're late to noticing. That's the outcome a standing watchlist exists to prevent.
5 · Does it serve anything you're trying to do?
The one everyone skips.
A format that reliably produces reach in your category may produce reach among people who will never buy from you. If it doesn't advance the goal you set, copying it buys you a metric and costs you a shoot day.
Reach in the wrong audience is not a cheaper version of reach in the right one. It's a different thing that looks the same in a dashboard.
The three responses
Having answered those, you have three sensible options and one bad one.
Ignore it. The correct answer more often than any other. Fluke, unrepeatable mechanism, wrong audience, or doesn't serve your goal. Write down why you're ignoring it, so the conversation doesn't restart in six weeks.
Watch it. The mechanism looks transferable but it's a single data point. Note it, set a reminder, and see whether their next attempt repeats it. Costs nothing.
Test it. The mechanism is transferable and it serves your goal. Make the cheapest credible version — not the flagship — and treat it as an experiment with a decision rule written in advance.
Copy it immediately at full scale. The bad one. You'll arrive late with a worse version of something whose mechanism you don't understand, and it will be visibly derivative to the audience that saw the original.
When it's your customer's content, not a competitor's
Worth a note, because it's the most under-exploited version of this.
Sometimes the breakout is in an adjacent category, or from a creator rather than a brand. Those are frequently more useful than competitor breakouts, because the mechanism arrives without the baggage of looking like a copy — nobody in your category will recognise where it came from.
Look sideways more than you look across.
Where Acumin fits
Watchlists score every competitor piece against that competitor's own distribution, so a breakout is flagged as unusual-for-them rather than by raw view count — which is the first step above, done continuously.
Because it runs on a schedule, you find out inside a week rather than when somebody forwards it a month later. That's most of the value: the difference between reacting to a pattern and reacting to a moment is almost entirely how early you saw it.
What it reads is public data — tier 1. It can't tell you whether a spike was paid, and it doesn't guess.
How to use this today
Take the last competitor piece someone forwarded you with "should we do this?"
Answer question 2 in one sentence: would this have worked if we had made it?
Most of the time that single question closes the file, and it's worth closing explicitly rather than leaving open as a background anxiety.
Related: Outliers, not virality is the method for defining unusual. Building a competitor watchlist is how to see these in the week they happen rather than the quarter after.