Reach, impressions, watch time and shares: reading social performance.

A social-media dashboard returns dozens of figures, and almost all change without you doing anything. Looking at them all daily creates the feeling of working on the project while, in practice, preventing you from seeing any trend clearly.

This guide arranges the six metrics found in almost every dashboard—reach, impressions, watch time, completion rate, saves and shares—according to the stage of the journey they describe, with each one’s most common reading error and suggested action.

What is measured and in what order

The six metrics describe a sequence: first the piece is shown, then someone decides to watch, stays for part of it, finishes or leaves, and finally decides whether it is worth returning to or sharing. Reading them in that order avoids the most common mistake: starting with the figure that rises most.

The table above fixes that sequence with one action per row. If you will look at only two numbers, choose those corresponding to the stage you are working on: if the content does not retain people, do not look at reach; if it retains people but nobody shares, do not change its length.

There is a practical rule: never compare metrics from different stages. Impressions are not assessed against completion rate, and reach is not explained by saves. Each stage has its own conditions and solutions.

Reach and impressions: two commonly confused numbers

Reach counts different people who saw the piece; impressions count how many times it was shown, including repeats. Someone seeing the same post three times contributes one to reach and three to impressions.

That is why reach tells you who arrives and impressions tell you how persistently the platform shows it. When impressions grow without reach, the piece is being shown repeatedly to the same people; when both grow, there is new distribution. Neither case alone tells you whether the content is good.

High reach can come from outside your usual audience: a trend, a share outside the network or paid distribution. Before reading it as growth, check whether those people did more than pass by: visited the profile, saved or clicked the link.

Watch time and completion rate

Watch time measures accumulated seconds; completion rate indicates how many reached the end. Both describe attention, and attention determines whether the platform offers the piece to more people.

YouTube presents a clear chain: thumbnail impressions generate clicks, those clicks produce watch time, and watch time determines later performance. If the chain breaks at the first link, packaging is the problem; if at the second, it is the content.

Completion rate also says something about length: low completion with high retention means the piece is too long for what it says; low completion with low retention means length is not the problem. Comparing different formats is another trap: CTR and retention vary with content type, audience and where the piece is shown.

Saves and shares: the intention behind the video

Saves and shares are the final part of the sequence and the one that weighs most in the medium term. A save signals reference value: someone wants to return. A share signals social value: someone takes the trouble to send it to another person.

The two have different characters. Useful technical content accumulates saves; content that moves people or makes them laugh accumulates shares. If your format does neither, the piece entertains and disappears: it passes the time rather than building something.

Always look at the ratio rather than the absolute figure. A piece with few impressions and proportionally many shares performs better than one with a million displays and no shares. If saves rise without shares, you have material to turn into a series; that signal most closely resembles an editorial direction.

The correct order for reading a report

Open the report and go through this sequence: impressions and reach to understand distribution; watch time and completion rate to see whether it held attention; saves and shares to see whether it left a mark; and finally profile visits and clicks to see whether it led towards the music.

Write one sentence per piece identifying the stage where it broke down. If people never reach the end, the problem is content; if they reach the end but do not share, it is the angle; if they share but nobody visits the profile, it is the call to action.

Repeat the exercise with pieces in the same format for two weeks. A homogeneous batch lets you draw conclusions; mixed pieces produce only unfair comparisons. This work fits with the guide to content metrics for DJs, which details the chain from retention and interaction to conversion into listening.

When the dashboard allows segmentation, add two breakdowns: visit source and publication time. The first tells you whether the piece spread within your audience or beyond it; the second, whether performance comes from the hour or day. Neither changes the content, but both change the interpretation of the same number, and that interpretation is what supports a decision.

Common social-dashboard mistakes

The first is confusing activity with progress: publishing more fixes neither low retention nor a badly placed link. The second is reading a daily figure as a trend; numbers fluctuate for reasons you cannot control, from the calendar to distribution luck.

The third is changing format whenever a piece fails, preventing accumulated learning. The fourth is looking only at the rising metric, a bias turning any dashboard into confirmation of what you already wanted to believe.

The fifth is forgetting the objective. If the content’s purpose is getting people to listen to the music, none of these six metrics is the destination: they are intermediate indicators. To discover whether the bridge to the music works, click as though you were a listener and check how many steps it takes to reach the track.

Comparing pieces fairly

Compare within the same format and with similar lengths: a forty-second live clip does not compete with a three-minute tutorial. Also compare over the same day window, because distribution in the first hours differs from the second week.

Replace the mean with the median once you have enough data: one viral piece distorts the account’s entire average and hides that the rest of the content is not working. Note alongside each comparison whether a campaign, collaboration or advertising was involved; otherwise, you will attribute the result to content when distribution changed.

Also set a decision threshold for each metric before looking at anything. For example, consider a piece a candidate for repetition only if it exceeds your retention median and records a save or share. Writing the threshold beforehand prevents moving the goalposts after seeing the result—the most elegant way to deceive yourself, and one frequently found in dashboards.

With the metrics in order, look at the other half of the dashboard: people who comment, visit the profile, click and convert.

Next step

Now comes the part connecting video to real work: comments, profile visits, CTR, conversion and attribution in Comments, CTR, conversion and attribution on social media.