How to Measure Reach

TL;DR

• Reach is the number of separate people who saw something, which makes it a count of humans rather than of deliveries.
• The formula is impressions divided by average frequency, so a million impressions at a frequency of 4 is 250,000 people.
• Impressions, views and reach measure different things, and the gap between the first and the last is usually four times or more.
• Platform reach figures resist being added together, since the same person appears on several of them.
• Summing three platforms at 270,000 can overstate true reach by 29% once the overlap is removed.
• Reach flattens while frequency climbs, so the back half of a long flight mostly buys repeats.
• We build Cruva, the AI-powered operating system for TikTok Shop affiliate growth, so creator reach lands in the same view as what it sold.

Reach is the number of distinct people who saw your content at least once.

That one word, distinct, is the whole metric, because every other number on a social dashboard counts events while reach tries to count humans.

Measuring it properly comes down to three things: knowing the formula, knowing what your platform means by the word, and resisting the urge to add platforms together.

What is the formula for reach?

Impressions divided by average frequency.

Reach = impressions ÷ average frequency

Frequency is the average number of times each person saw the content, so a campaign serving a million impressions to an audience that saw it four times each has reached 250,000 people.

Rearranged, the same relationship gives you frequency when the platform reports reach instead, which is frequency = impressions ÷ reach.

Most platforms report two of the three numbers and leave you to derive the third, so knowing the triangle keeps you moving in almost every case.

How do reach, impressions and views differ?

They count deliveries, plays and people, in that order.

• Impressions count every time the content was served, including repeats to the same person.
• Views count every time somebody actually watched, on whatever definition of a view that platform uses.
• Reach counts the people, each one once.

The spread matters because these three get quoted interchangeably in reports, and the first number is usually four times the last.

The definition of a view is the softest of the three, since platforms have historically counted anything from a fraction of a second upward, so read the definition before you compare two platforms on it.

Where a brand is making an awareness case rather than a sales case, reach is the number that belongs in the deck, for the reasons set out in brand awareness.

Why can you avoid adding platform reach together?

Because the same person shows up in more than one number.

Each platform measures reach inside its own walls, so a person who follows you on TikTok and Instagram is counted by both, and the deduplication falls to you.

Take three platforms reporting 120,000, 90,000 and 60,000.

The overlap figure is the one everybody is missing, which is the honest problem, since measuring it properly needs a panel or a matched survey rather than a dashboard.

A workable practice is to report platform reach separately, state that the figures overlap, and give a deduplicated estimate with the assumed overlap written beside it.

That is less satisfying than one big number and considerably harder to be embarrassed by later.

What happens to reach over a longer campaign?

It flattens, while frequency keeps climbing.

A campaign pointed at a fixed audience picks up most of its reach early, because the easiest people to reach get reached first.

• Day 1 reaches 50,000 at a frequency of 1.2.
• Day 7 reaches 150,000 at 2.0.
• Day 14 reaches 200,000 at 3.5.
• Day 30 reaches 220,000 at 5.5.

The last sixteen days added 20,000 people and nearly doubled frequency, so the second half of that flight was mostly buying repeats.

That shape is the argument for widening the audience or refreshing the creative rather than extending the flight, and it is why a roster of small creators often out-reaches one large placement, as covered in nano influencers.

It is also the reason an affiliate programme that keeps adding creators behaves differently from an ad flight that keeps running, which is part of the case in affiliate programme ROI.

Reading the curve is also how you avoid mistaking frequency for growth, since a rising impression count against flat reach means the same people are seeing it again.

How do you measure reach when the platform hides it?

Estimate it from what you can see, and say that you estimated it.

Three routes work, in descending order of confidence.

• Divide impressions by frequency, where both are reported, which is exact.
• Multiply followers by an observed reach share, where a creator's recent posts show what proportion of their audience a typical post touches.
• Use views as a ceiling, since reach stays at or below the number of plays and usually sits well below it.

The second route is the one most useful for creator campaigns, because you can read it off a handful of recent posts before you book anybody.

That estimate is usually enough to sanity check a quoted rate, which is the arithmetic behind how much to charge for a TikTok post.

Whatever you use, write the method beside the number, as a reach figure with zero method attached is the easiest thing in marketing to argue with.

Reach also stops short of telling you whether any of it worked, which is where ROAS and the delayed demand covered in the halo effect come in.

What should you report alongside reach?

Three numbers, so the figure means something.

Frequency tells the reader how concentrated the delivery was, and a reach number quoted alone can describe almost any campaign.

Cost per thousand reached turns reach into an efficiency measure you can compare across channels, which is a different calculation from the usual cost per thousand impressions.

What the reach produced closes the loop, whether that is branded search, site visits or sales, and the way creator programmes get valued on both halves is set out in media value.

Reported together these four lines fit on one slide and survive questions, which is more than most reach numbers manage.

Counting People Rather Than Deliveries

Reach is the count of distinct people, so divide impressions by average frequency to get it.

Keep impressions, views and reach apart, since the first is usually four times the last.

Resist adding platforms together, because three numbers summing to 270,000 can mean 210,000 real people.

Then report frequency beside it, or the number says very little.

FAQs

What is reach in marketing?

The number of distinct people who saw your content at least once. It counts humans rather than deliveries, which is what separates it from impressions.

How do you calculate reach?

Divide impressions by average frequency. A million impressions at a frequency of 4 means 250,000 people saw the campaign.

What is the difference between reach and impressions?

Impressions count every delivery including repeats to the same person, while reach counts each person once. The gap between them is usually four times or more.

What is the difference between reach and views?

Views count plays, so one person can produce several. Reach counts people. A view definition also varies between platforms, so check it before comparing.

Can I add reach across platforms?

Only with a caveat. The same person appears on several platforms, so three figures summing to 270,000 might represent 210,000 real people once the overlap is removed.

What is a good frequency?

Enough for memory and little enough to avoid waste, which in practice means watching the point where extra days stop adding reach. In the example here, days 14 to 30 added 20,000 people while frequency rose from 3.5 to 5.5.

How do I measure reach if the platform hides it?

Divide impressions by frequency where both appear, or multiply a creator's followers by the share of them a typical recent post touches. Write the method beside the number.

What should I report with reach?

Frequency, cost per thousand people reached, and whatever the reach produced downstream. Reach alone invites the question of how concentrated it was.