Instagram by the numbers — users, time spent and who is on it

You check it twelve times a day. Each time for six minutes. You do not think about it. No one does. But the numbers behind Instagram tell you something about the thing you are using. How big it is. Who else is on it. How much time they give it. And what Meta chooses to count versus what it hides.

The clearest numbers come from Meta itself. The company reports user counts to investors. It names milestones in blog posts. Some of it sits in earnings transcripts. Some sits in research paid for by someone else. The rest is guesswork dressed as fact.

Here is what the primary sources actually say. What they count. How they measure it. And where the trail runs cold.

Every number below is linked to where it came from. Every quote is verbatim. One section lists what we could not verify.

Three billion monthly users, two billion daily

Instagram crossed three billion monthly active users in September 2025. That makes it the fourth platform to hit that mark. Facebook, YouTube and WhatsApp got there first.

We could not read Meta's own announcement of the milestone. The company appears not to have published one on its news site. Trade press reported it. CNBC and others quoted Mark Zuckerberg announcing it. But we could not open a primary source carrying his words on that specific claim.

What we could read is the Q2 2026 earnings call. Zuckerberg said this:

Instagram reached 2 billion daily actives. — Mark Zuckerberg, Meta Q2 2026 earnings call, 7 August 2026

That is daily active users. People who open the app at least once in a day. The three billion number counts monthly actives instead. Anyone who used it in a month. So roughly two thirds of Instagram's monthly users come back daily. That is a high retention rate. It means the app is not just installed. It is used.

The same call gave one more number. Time spent:

On Instagram, global time spent this quarter grew double digits year-over-year. — Mark Zuckerberg, Meta Q2 2026 earnings call, 7 August 2026

Double digits means at least ten percent growth. Year over year. That is usage rising, not plateauing. But Zuckerberg gave no absolute number. No "users spend X minutes per day". Meta does not publish that in its own documents.

Third parties do. DataReportal published a global digital report in early 2026. It names its sources. GWI for survey data. Similarweb for app tracking. The numbers it gives:

• 1 hour and 13 minutes per day on Instagram's Android app
• 12.3 times opened per day, on average
• 5 minutes and 56 seconds per session

Those are averages. From Similarweb's app tracking. The report does not give a sample size for the tracking pool. It does name the tool: Similarweb's App Intelligence product. That tracks app usage across devices. How many devices is not stated.

So the 73 minutes a day is a measured average. Not a Meta claim. It comes from a vendor that tracks apps. The method is disclosed. The sample size is not.

Half of American adults use it

Pew Research Center surveyed 5,022 U.S. adults between 5 February and 18 June 2025. Fifty percent said they use Instagram. That is one in two American adults.

The split by age is steep. Eighty percent of adults aged 18 to 29 use it. Nineteen percent of those 65 and older do. Instagram is a young platform in an aging country.

Here is Pew's full breakdown:

By age:

• 18-29: 80%
• 65+: 19%

Pew did not publish separate numbers for the 30-49 and 50-64 groups in the source we read. It gave the endpoints.

By gender:

• Women: 55%
• Men: 44%

More women than men. An eleven-point gap.

By race and ethnicity:

• Hispanic: 62%
• Asian (English speakers): 58%
• Black: 54%
• White: 45%

Hispanic adults use Instagram at the highest rate. White adults at the lowest. That is a seventeen-point spread.

By education:

• College graduate: 58%
• Some college: 53%
• High school or less: 41%

The more education someone has, the more likely they are on Instagram. Pew names Instagram explicitly as one of the platforms where this pattern holds.

By household income:

• $100,000 or more: 60%
• $70,000-$99,999: 54%
• $30,000-$69,999: 46%
• Less than $30,000: 41%

Higher income, higher usage. A nineteen-point gap from top to bottom.

By community type:

• Urban: 55%
• Suburban: 54%
• Rural: 37%

City and suburban users look the same. Rural users trail by eighteen points.

Pew's survey used address-based sampling. It reached people by web, mail and phone. The margin of error for the full sample is plus or minus 1.7 percentage points. For subgroups the margin is larger. Pew does not report it for each one.

One thing the survey measures: who is on the platform. Not how they use it. Fifty percent of American adults have an account. That says nothing about how often they open it. Or for how long. Or whether they post or just scroll.

What Meta reports to investors versus what it hides

Meta's quarterly earnings reports give one top-line number. Family daily active people. That counts everyone who used Facebook, Instagram, Messenger or WhatsApp on a given day. In Q2 2026 that number was 3.60 billion.

Three point six billion across four apps. Meta does not break it out. You cannot tell from the filing how many of those billions touched Instagram. Or Facebook. Or both. The company aggregates them into one metric. It calls it the Family of Apps.

Instagram's two billion daily actives came from the earnings call. Not from the written report. Zuckerberg named it while talking to analysts. It does not appear in the press release Meta filed with the SEC.

That pattern holds across quarters. Meta reports Family numbers in the filings. It names Instagram numbers in calls and blog posts. Sometimes. Not every quarter. When it does, the number is a milestone. Two billion daily actives is a round number. It makes news. A 3 percent gain does not.

So investors get aggregated metrics. The public gets milestones. What stays hidden: Instagram's quarter-to-quarter growth rate. Its revenue as a separate line. How many of its users overlap with Facebook. Meta does not publish any of that.

One claim we see repeated: Instagram has 350 million business accounts. That number comes from trade press in 2025. We could not trace it to a Meta document. It may be real. We cannot verify it.

What time spent actually measures

Seventy-three minutes a day sounds like a lot. It is also an average. Which means half the users spend less. The other half spend more. Averages compress the spread. The median would tell you more. Similarweb does not publish one.

The 12.3 daily sessions number matters more. Twelve times a day means Instagram is not a once-in-the-evening thing. It is a checking behavior. You open it. You look. You close it. Six minutes later you do it again.

That is the product working as designed. Every feature on Instagram is built to bring you back. Stories expire. Reels autoplay. The red dot on the notifications tab pulls you in. Twelve sessions a day is not an accident. It is the goal.

DataReportal's report names a second metric. Session duration. Five minutes fifty-six seconds. That is how long one visit lasts, on average. Multiply it by 12.3 sessions and you get 73 minutes. The math checks.

What the numbers do not tell you: how much of that time is active versus passive. Watching a reel is passive. Writing a comment is active. Scrolling while half-awake on the train is something else. The tracking counts all of it the same. App open equals time spent.

One more thing those numbers miss. They come from Android tracking. Similarweb's tool measures Android apps. Not iOS. Not the mobile web. Not desktop. So the 73 minutes is Android only. iPhone users may spend more or less. The report does not say.

The advertising reach numbers, and why they are not user counts

Meta publishes advertising reach through its Ads Manager. That tool shows you how many people you can target with an ad. Broken down by country, age, gender and interest. Third parties pull that data. NapoleonCat does. DataReportal does. They republish it as user demographics.

Here is what that number actually counts. It is the size of the audience you can buy ads against. Meta calls it potential reach. The word potential matters.

Potential reach includes everyone eligible to see an ad. That includes people who have not opened Instagram in weeks. It includes multiple accounts by one person. It includes accounts that see ads but never post. It counts reach, not usage.

So when a statistics page says "Instagram has X users aged 18-24", it is usually quoting ad reach. Not active users. The two are not the same. We cannot give you those numbers here. The reason: we could not open the advertising tools or the NapoleonCat breakdowns directly. The pages redirected or blocked automated access.

We know the numbers exist. Meta publishes them. But reproducing them requires accessing the Ads Manager API yourself. Or trusting a vendor who did. We did neither. So we report nothing on global demographics by age and gender. The claim would rest on a source we could not verify.

What we could not determine

Five things we tried to find and did not.

The primary source for three billion monthly users. Trade press reported it in September 2025. They quoted Zuckerberg. We searched Meta's news site and Instagram's blog. We found secondary accounts. No direct post from Meta or Instagram carrying the announcement. It may exist. We could not open it.

Instagram-specific revenue. Meta does not break out Instagram's ad revenue in its filings. It reports one number for the Family of Apps. Analysts estimate Instagram's share. We found no primary source.

iOS time spent. Similarweb's Android tracking gave 73 minutes a day. No equivalent number for iPhone users appeared in any source we read.

The sample size for Similarweb's app tracking. DataReportal names Similarweb as the source. It describes the tool. It does not give a sample size. How many devices Similarweb tracks is not stated.

Advertising reach demographics by age and gender. Those numbers exist in Meta's Ads Manager. We could not access the tool directly. Third-party sites that republish the data blocked our access or redirected. So we verified no global age-and-gender breakdown.

Where the useful numbers live

The best numbers come from three places. Meta's earnings calls. Pew's surveys. And vendor research with a disclosed method.

Earnings calls are primary. Zuckerberg speaks to investors under legal obligation. What he says there is as close to truth as you get from a company. The transcripts are public. They sit on Meta's investor site. Sometimes paywalled. Sometimes not. The Motley Fool republishes them. That is where we read the Q2 2026 call.

Pew Research is the gold standard for U.S. demographics. It surveys thousands of adults. It discloses the method. It reports the margin of error. It releases the full questionnaire. If Pew says fifty percent of American adults use Instagram, that claim rests on a sample you can check.

Vendor research is weaker but usable. DataReportal names GWI and Similarweb. It says what each one measures. The limit: you cannot see the raw data. You trust the vendor's summary. That is one step removed. But it is disclosed. That beats a marketing blog with no source at all.

Everything else is noise. The "2026 Instagram statistics" round-ups copy each other. They cite no primary source. They mix ad reach with active users. They report three-year-old numbers as current. And they rank in search because they answer the question people type. None of it is reliable.

The number that matters is the one you cannot see

Instagram has three billion monthly users. You are one of them. Or you are not. That number tells you the platform is huge. It does not tell you whether your posts get seen.

The reach number you care about is not in this post. It is in your insights. How many accounts your last post reached. That is the number that moves. It is also the number Meta does not explain. The algorithm decides it. The algorithm does not publish its weights.

So Instagram tells you the total. Three billion people. Two billion daily. It tells you the average. Seventy-three minutes a day. What it will not tell you: why your post reached 300 people instead of 3,000. That stays in the black box.

The box is bigger than one number. Meta does not publish the weights in its ranking models. It does not publish any curve showing how engagement in the first hour affects reach in the next twelve. Those numbers would settle most of the arguments people have about the feed. They do not exist in any document we could find.

You now know the shape of the platform. How big. How often used. Who is on it. That is the context. The outcomes happen one post at a time. And those, Meta does not aggregate.

Read next in Platforms & Industry

This category takes the same approach to the rest of the field: read what each company publishes, name the source, and mark the gaps. These extend what is above.

• Every platform's numbers in one place — the same method applied across the field, source by source.
• What X will say about its size, and what it will not — the first audited user count in four years, and the filing it arrived in.
• YouTube's revenue and the claims you cannot check — what Alphabet files, and what the press release adds on top.
• LinkedIn's 1.3 billion members against its 310 million users — the gap between holding an account and opening the app.
• Where the time is going now — the move into private sharing, read off Instagram's own ranking signals.
• How Instagram got here — fifteen years of changes, each one dated and sourced.

How we did this

Scope: this post reads published sources and reports what they say. No original data collection. No experiment.

How we read them: we opened each source directly on 18 August 2026. We copied quotes verbatim from the pages. Some sources could not be opened. The Meta earnings PDF returned encoded data we could not read. Several CNBC and Statista URLs returned 403 errors or redirect loops. NapoleonCat's statistics pages redirected or blocked access. Where we could not open a source, we say so. We cite nothing we did not read.

The Pew data comes from its published report dated 20 November 2025, covering a survey conducted 5 February to 18 June 2025. The DataReportal numbers come from its Digital 2026 Global Overview Report, which discloses GWI and Similarweb as sources and states it was written without generative AI. The Zuckerberg quotes come from the Motley Fool's transcript of the Meta Q2 2026 earnings call, published 7 August 2026.

Limits: Most sources are company statements or vendor research. Meta decides what to publish about itself. Pew surveys U.S. adults only. It says nothing about the rest of the world. Similarweb tracks Android apps, not the full user base. DataReportal aggregates others' research. It is two steps from the measurement.

Sources

1. Mark Zuckerberg, Meta Q2 2026 earnings call. Transcript published by The Motley Fool, 7 August 2026. Two quotations: "Instagram reached 2 billion daily actives" and "On Instagram, global time spent this quarter grew double digits year-over-year." Retrieved 18 August 2026 from https://www.fool.com/earnings/call-transcripts/2026/08/07/meta-meta-q2-2026-earnings-call-transcript/.

1. Pew Research Center, Americans' Social Media Use 2025, 20 November 2025. Survey conducted 5 February to 18 June 2025. Sample: 5,022 U.S. adults. Methodology: address-based sampling, multimode (web, mail, phone). Margin of error plus or minus 1.7 percentage points for the full sample. Source of all U.S. demographic breakdowns. Retrieved 18 August 2026.

1. DataReportal, Digital 2026 Global Overview Report, January 2026. Time-spent and session data attributed to Similarweb's App Intelligence product. User survey data attributed to GWI's Q2 2025 survey of online adults aged 16+. The report states: "All of this analysis has been written by Simon Kemp without the use of generative AI" and names data partners as "GWI, Statista, GSMA Intelligence, Similarweb, Semrush, Ookla, SocialInsider, and Skai". Three statistics cited: 1 hour 13 minutes daily on Instagram's Android app, 12.3 sessions per day, 5 minutes 56 seconds per session. Retrieved 18 August 2026.

1. Meta Platforms, Inc., Q2 2026 earnings press release, 7 August 2026. Family daily active people: 3.60 billion. Retrieved 18 August 2026 from Meta's investor site.

1. About Instagram blog and Meta newsroom. Searched 18 August 2026 for the three billion user announcement. Secondary sources (trade press) reported it in September 2025, attributing it to Zuckerberg. We found no primary post on Meta or Instagram's official sites carrying the announcement directly.

Changelog

• 2026-08-18: First version drafted.