The question always comes after the first campaign: did it work?
In Italy it is harder than elsewhere for a simple reason. A significant share of sales does not close online, and what does not close online is not visible in a dashboard.
What your dashboard will never see
Someone sees the video, goes to the shop and buys. Someone sees the video and writes in a messaging app to ask whether the product is in stock. Someone sees the video and mentions it to a reseller.
In all three cases the sale exists and the measurement tool attributes it to nothing. Many brands then conclude the video produced nothing, and stop at the precise moment it was starting to work.
The problem is not the video, it is that a channel is being measured with another channel's instrument.
The three signals to follow online, and nothing else
Three-second retention. It tells you whether the hook works. It is the only measure that judges the start of the video, and the start decides everything else.
Click-through rate. It tells you whether the video makes people want to look. A much-watched but little-clicked video entertains, it does not sell.
Cost per purchase, when the purchase is online. It is the only measure that counts for paid advertising, and it makes every vanity metric useless.
Mentions, shares and comments are pleasant. They prove nothing and should never decide whether you continue.
What is measured offline, concretely
The volume of incoming messages, week by week, before and during the campaign. It is the fastest signal in Italy, because messaging is the first reflex.
Calls and shop visits over the same period, with the question above.
And a code spoken aloud in the video, to be mentioned in store or in the message. A spoken code works where a link does not, and it gives an imperfect but real attribution.
The measurement window
A week is not enough. A published video does not produce its effect the same day, especially when the decision involves a trip or a discussion at home.
Look over four to six weeks, and always compare with the same period last year rather than with last month. Commercial seasons distort everything, and a dip in August says nothing about the quality of your content.
What to decide before shooting
What you will measure, and from what threshold you continue. Written beforehand, never after: a measure chosen afterwards always ends up confirming what you hoped.
And one variable at a time. If you change the creator, the hook and the audience together, you will know something changed without ever knowing what.
What to look at, in order
| Indicator | What it tells you | When to read it |
|---|---|---|
| Share still there at 3 s | The strength of the hook | The next day |
| Clicks per thousand views | The real interest | At three days |
| Cost per order | The profitability | At two weeks |
What is noted before publishing
The goal, the format, the hook used and the landing page: without those four lines, no figure explains itself.
What is never compared
Two videos posted on different dates, to different audiences, with different budgets.
The costliest mistake
Judging a video on its view count. A video seen by a thousand people who were hesitating beats a video seen by a hundred thousand who will never buy.
That is especially true for a company whose market is regional: the useful audience is small, and an audience metric will always call it disappointing while the order book says the opposite.
