Reach vs impressions: the definitions, and what the gap tells you

Two numbers sit next to each other in every analytics screen, one is always bigger, and almost nobody can say what the gap means. Then a report goes to a leadership team with one of them in it, chosen more or less at random, and a quarter of strategy gets decided on a metric nobody defined.

This page settles reach vs impressions using the platforms' own definitions, shows what the gap between them actually tells you, and explains why neither is visible without account access. Written for founders, marketing leads and anyone who has to report on social work.

TL;DR

The definitions, from the platforms

LinkedIn defines impressions as the number of times a post was displayed on screen, and adds that the number is an estimate and may not be precise. It defines unique viewers as the number of members a post was displayed to at least once. Its own example is exact: a post shown to one member five times counts as five impressions and one unique viewer.

Reach is the same idea as unique viewers, under a different label. Networks use both words, which is most of the reason the terms feel slippery. Meta uses reach, LinkedIn uses unique viewers, and both mean people rather than displays.

Impressions count screens. Reach counts people. Everything else about the two follows from that one line.

What the gap between them means

Divide impressions by reach and you get displays per person. Close to one means each viewer saw the post once, which is normal for a feed post. Well above one means the same people saw it repeatedly, which usually means the post kept resurfacing to an audience that already knew it.

That ratio is more useful than either number alone, and almost nobody reports it. A post with high impressions and low reach did not travel. It circled.

Why both numbers move without you doing anything

Definitions change. LinkedIn states that since April 2024 the impression count on a post includes impressions on reposts of that content by other members, which by itself can raise the number without any change in your work.

Platforms also add deliberate noise for privacy. LinkedIn says it introduces a small amount of noise into the demographics of unique viewers to prevent visibility into exact viewer behaviour, and shows those results as percentages rather than counts.

So a month over month comparison of impressions can reflect a definition change, a repost by someone with a large audience, or rounding. This is why judging a team on reach is unfair: the number moves for reasons nobody in the room controls.

Neither number is visible from outside the account

This is the practical consequence most guides skip. Reach and impressions are shown to the account owner, inside their analytics. A visitor cannot see them, and neither can a tool that has not been given access to the account.

Meta's developer documentation is explicit that reading insights such as impressions and reach requires an access token from the Instagram business user. Where the public line sits on every network is mapped in public social media metrics.

For a team of ten people publishing under their own names, that means reach based reporting requires ten connections, ten approvals, and a new one every time somebody joins. It is the reason most team programs report on something else.

What to report instead, by situation

• Optimising one account you own. Report reach, and read the impressions per person ratio next to it. This is what native analytics are for.
• Running a team program. Report published cadence and audience growth. Both are visible without access, and both are things people control. See follower growth rate.
• Reporting to a board or a client. Report the leading indicator, which is consistency, next to the lagging one, which is audience. Reach in a quarterly deck invites a question nobody can answer honestly.
• Comparing yourself to competitors. You cannot see their reach either, so compare what is public. See track competitor social media.

Views, the third number nobody defines

Video adds a third word, and it is the loosest of the three. A view is a play that passed some threshold, and the threshold differs by network and changes over time. It is not reach, because one person can play a video several times.

The practical rule is to never compare views across networks. A view on one platform and a view on another are different measurements wearing the same label, and a report that adds them together is producing a number that describes nothing.

How to compare two months honestly

• Compare the same metric, from the same screen. Native analytics and third party tools count differently, so pick one and stay there.
• Compare periods of the same length. Thirty days against thirty days, never a calendar month against four weeks.
• Divide by posts published. Reach per post separates a better month from a busier one, which is the question you actually have.
• Note platform changes. A definition change explains more month over month jumps than any content decision.

That last point is why consistency belongs next to any audience metric in a report. It is the one line nobody can attribute to a platform update, as social media posting consistency argues.

Engagement rate: over reach, or over followers

Engagement rate needs a denominator, and the two candidates answer different questions. Divided by reach, it measures how well a post landed with the people who actually saw it. Divided by followers, it measures how far the post travelled inside the audience the account already had.

The first is the better quality signal and it needs the account. The second is computable from public data, which is why it is the one that survives in team reports and in competitor benchmarks. Whichever you pick, write the denominator next to the number, because one engagement rate can be double another and mean less.

Three mistakes these two numbers cause

• Summing impressions across networks. Each platform counts a display differently, so the total describes no real event. Give each network its own line.
• Reading a drop as a content problem. Distribution moves on platform decisions. Check whether the definition changed before you change the strategy.
• Setting a reach target for a team. Nobody in the room controls it, so the target is hit by luck or missed by luck, and neither outcome teaches anyone anything.

One pattern sits behind all three. These are outcome metrics, and an outcome metric makes a poor management tool, because the person being managed cannot move it directly.

Explaining the two numbers to someone who does not care

Most people who read your report do not want a definition, they want to know whether the work is going anywhere. Two sentences do the job better than a chart.

Say that impressions count screens and reach counts people, so impressions is always the bigger of the two. Then say which one you are reporting and why, and stop. The moment you offer both without a reason, the meeting turns into a discussion about which number is more flattering.

Pick one number, name it, and keep it for a year. A metric that changes every quarter cannot show a trend, which was the entire point of reporting it.

If you only have room for one line, report reach per post rather than reach. It separates a better month from a busier one, and that is the question the person reading actually has.

Both of these serve as the denominator of the same rate, and the choice between them changes the answer by an order of magnitude, as social media engagement sets out.

How Groowth handles this

Groowth never shows reach or impressions, on any screen, for any account. It cannot see them without a login it does not ask for, and a product that displays a number it cannot source is a product you cannot trust with the numbers it can.

What it does show is what a visitor can see and what a team controls: posts published against a cadence, streaks, follower growth, and public engagement on each post. Three accounts free, no card, no connection to any network.

Frequently asked questions