TL;DR, Quick Answer
8 min readAccounts engaged is the number of unique accounts that interacted with your content in a period. Interactions is the number of actions. One account that likes, saves and comments adds 1 to accounts engaged and 3 to interactions. Instagram labels the number estimated and in development, the API field is accounts_engaged, and there is no per-post version of it in the API.
What does accounts engaged mean on Instagram?
Meta's accounts engaged Instagram figure is the number of unique accounts that interacted with your content during a period, not the number of interactions those accounts produced. One person who likes a post, saves it and leaves a comment adds 1 to accounts engaged. That same person adds 3 to Interactions.
Instagram's Help Center puts it this way: "The number of unique accounts that have interacted with your content and demographic information on the accounts you've engaged, including top countries, top cities, top age ranges and gender breakdown."
Meta's API reference for the same number reads differently: "The number of accounts that have interacted with your content, including in ads. Content includes posts, stories, reels, videos and live videos. Interactions can include actions such as likes, saves, comments, shares or replies."
Two words separate those definitions. The app says unique and the API does not. The API says including in ads and the app does not. Read them together and you get the working definition: a deduplicated count of accounts, covering organic and boosted delivery, reset for each period you request.
The period matters more than most reporting decks admit. At the API level accounts_engaged runs on period=day, so an account that engages on Monday and again on Thursday counts once in each day's number. Summing daily accounts engaged across a month does not give you monthly accounts engaged. It gives you something larger with no name.
How is accounts engaged different from interactions?
Accounts engaged counts who, Interactions counts what. Instagram defines Interactions as "The number of actions people take when they engage with your content, such as likes, comments, saves and shares." Nothing in that sentence deduplicates by account, and nothing in the accounts engaged definition adds up actions.
Put a real post through both. Say 40 accounts touch it: 30 like it once, 8 like and save, 2 like, save, comment and share. Accounts engaged is 40. Interactions is 30 + 16 + 8, so 54. Divide one by the other and your average engaged account took 1.35 actions. That ratio is the only thing either metric tells you about depth, which is why both belong in the report.
| App label | Counts | Deduplicated | API metric | Meta's disclaimer |
|---|---|---|---|---|
| Accounts engaged | Accounts | Yes, once per period | accounts_engaged | Estimated |
| Interactions | Actions | No | total_interactions | In development at media level |
| Viewers | Accounts | Yes, once per period | reach | Estimated and in development |
| Views | Plays and displays | No | views | In development |
The pairing is symmetrical. Viewers is to Views what Accounts engaged is to Interactions: the account-level number sitting under the action-level number.

Is accounts engaged an estimated metric?
Yes, and as of 10 September 2026 the disclaimer is still on Meta's own page. The Help Center article on Instagram insights attaches this note to both the Viewers and Accounts engaged entries: "These metrics are estimated and in-development." The API reference ends the accounts_engaged description with "This metric is estimated."
Meta explains what that label means on its metrics labeling page. "Estimated metrics are derived through statistical sampling or modeling, not a straight count." The same page gives the reason this particular metric qualifies: "Our metrics that count Meta Accounts are sampled because it takes a large amount of data to calculate them." Counting accounts is expensive, so Meta samples.
The in development half carries a sharper warning. Meta writes that "In development metrics are measurements we're still testing. We're working out the best way to measure them and may make adjustments until they're accurate," and then states outright that businesses should not use in development metrics for "Historical comparison of advertising performance" or "Strategic planning or important business decisions." That is Meta telling you not to put this number in a year-over-year chart.
There is a gap here worth naming. Meta's metrics labeling page publishes lists of every metric carrying each disclaimer. Viewers appears on both lists. Interactions appears on the in development list. Accounts engaged appears on neither, even though the Instagram Help Center and the API reference both label it. The disclaimer is real. The index of disclaimers has not caught up.
Why does my dashboard show different numbers from the Instagram app?
Because Instagram's app labels and Meta's API field names drifted apart, and most tools read the API. The app was relabelled. The field names were not.
| Instagram app label | API metric | Level the metric exists at |
|---|---|---|
| Accounts engaged | accounts_engaged | Account only |
| Interactions | total_interactions | Account and media |
| Viewers | reach | Account and media |
| Views | views | Account and media |
Viewers is the clearest case. Open the app and the metric is called Viewers. Ask the API for reach and the sample response in Meta's own reference comes back with "title": "Accounts reached". Same number, two names, and a dashboard that faithfully prints the API title will show a heading no user of the app recognises.
AdaptlyPost
Start 7-Day FREE Trial
All-platform analytics
Social Inbox
AI-powered assistant
The Instagram Platform changelog carries entries from February 2018 through 22 June 2026 and records none of this. It documents deprecations, new fields and permission renames in detail. It has no entry for the app-side label change. If you have ever wondered why tool vendors ship inconsistent Instagram column headers, that silence is the answer.
Meta's own reference is not internally consistent either. The metrics table for reach says "This metric is estimated." The sample response further down the same page returns a description ending "This metric is estimated and in development." Two disclaimers, one page.
Can I get accounts engaged for a single post?
Not from the API. The Instagram app shows Accounts engaged when you open insights on an individual post, story, reel or live video. The media insights endpoint has no accounts_engaged metric at all. Its list runs comments, crossposted_views, facebook_views, follows, likes, link_clicks, navigation, profile_activity, profile_visits, reach, reels_skip_rate, replies, reposts, saved, shares, total_interactions and views, plus the aggregated total_likes, total_comments and total_views. No account-level engagement count anywhere in it.
The follower split has the same problem. The Help Center says that "For posts, stories and Instagram Live, you can see the Accounts engaged metric broken down into followers and non-followers." In the API, accounts_engaged lists no compatible breakdown. Request one anyway and you get back the least helpful error string Meta publishes: "An unknown error has occurred."
Three more field-name traps sit in the same pair of documents.
Account-level insights call the metric saves. Media-level insights call it saved. Same concept, one letter apart, no alias.
Account-level total_interactions is "The total number of post interactions, story interactions, reels interactions, video interactions and live video interactions, including any interactions on boosted content." Media-level metrics are explicitly the opposite: comments, likes, views and total_interactions there "report organic interaction metrics only; interactions on ads containing a media object are not counted." Identical field name, opposite treatment of paid.
The views row in the account-level metrics table lists follower_type as a breakdown. The breakdown section on that same page defines follow_type. One of those two spellings returns data.

What limits apply to accounts engaged data?
The app caps insights at "a preset or custom timeframe, within the past 90 days." Anything older than 90 days is not there to export.
| Limit | Value | Where it applies |
|---|---|---|
| Data delay | Up to 48 hours | Account and media insights |
| Minimum followers for insights | 100 | Account insights |
| Minimum engagements for demographics | 100 in the timeframe | engaged_audience_demographics |
| Demographic detail returned | Top 45 performers | Demographic metrics |
| Media metric retention | 2 years | Media insights |
| Story metric availability | 24 hours | Story media |
| Story insights floor | Fewer than 5 returns error code 10 | Story media |
That story floor is a literal string worth grepping your logs for: (#10) Not enough viewers for the media to show insights. It is not a failure, it is a privacy threshold.
One behaviour catches anyone writing their own aggregation. When insights data does not exist or is unavailable, Meta returns "an empty data set instead of 0 for individual metrics." Code that assumes a numeric zero will crash or drop the day without saying so.
The request itself:
GET https://graph.facebook.com/v26.0/<IG_USER_ID>/insights
?metric=accounts_engaged
&period=day
&metric_type=total_value
&access_token=<TOKEN>
Note that impressions is gone. Meta deprecated it for v22.0 and for all versions on 21 April 2025, and at media level it is deprecated for anything created after 2 July 2024. Old reporting templates that still request it are requesting nothing.
How should you report accounts engaged?
Report it beside Interactions and never in place of it. The two answer different questions, and one of them is a sampled estimate. Label every column with its source, app or API, so nobody averages two figures that were never meant to reconcile. When the app and your dashboard disagree by a few percent, the sampling in Meta's own definition already accounts for it.
AdaptlyPost
Start 7-Day FREE Trial
All-platform analytics
Social Inbox
AI-powered assistant
Pulling the number on a schedule instead of screenshotting it once a month is what makes the daily period usable, and it is the job social media analytics in adaptlypost handles alongside the other networks you publish to. The Instagram analytics tool view sits next to the Instagram scheduling tool that produced the posts you are measuring.
Frequently Asked Questions
Does accounts engaged count the same person twice?
Not within one period. Accounts engaged deduplicates by account, so an account that likes, saves and comments on the same post counts once. Across periods it counts again. An account that engages on two separate days appears in both daily figures, which is why adding daily values together does not produce a valid weekly or monthly number.
Does accounts engaged include ads and boosted posts?
The API definition says yes, in the words "including in ads. Content includes posts, stories, reels, videos and live videos." The Help Center definition of Accounts engaged does not mention ads. Treat the API figure as the wider one covering both organic and boosted delivery.
Which API field returns accounts engaged?
The field is accounts_engaged on the account-level insights endpoint, requested with period=day and metric_type=total_value. It requires the instagram_business_manage_insights permission with Instagram Login, or instagram_manage_insights plus instagram_basic and pages_read_engagement with Facebook Login.
Why does the API say Accounts reached when the app says Viewers?
The app label changed and the API did not. The metric is still reach, and Meta's own sample response returns "title": "Accounts reached". The Instagram Platform changelog records no entry for the label change, so tools that print the API title keep showing Accounts reached to people who only ever see Viewers.
Can I split accounts engaged into followers and non-followers?
In the app, yes, for posts, stories and Instagram Live. Through the API, no. The accounts_engaged metric lists no compatible breakdown, and requesting one returns "An unknown error has occurred." The reach metric does support a follow_type breakdown, so the follower split is available for viewers but not for engaged accounts.
Why did the impressions metric stop working?
Meta deprecated impressions for v22.0 and for all API versions on 21 April 2025, at both account and media level. At media level it survives only for media created on or before 1 July 2024. The replacement is views, which carries the in development disclaimer and supports breakdowns by media product type.
How many followers do I need before Instagram shows accounts engaged?
Instagram only turns on account-level insights once a profile passes 100 followers, and Accounts engaged lives inside that account view. Profiles below that count never see the metric, no matter how much engagement individual posts pull in.
How long does it take for accounts engaged data to update?
Account insights and media insights both carry a data delay of up to 48 hours. A number that looks low right after publishing can still climb during those two days. Check the figures again once that window passes instead of treating a fresh pull as final.
How far back can I pull accounts engaged data from the app?
The app caps insights at a preset or custom timeframe within the past 90 days. Anything older than that is not there to export from the app. Getting your own history past 90 days means pulling and storing the daily numbers before they age out.
What happens when there is no accounts engaged data for a day?
Meta returns an empty data set instead of a 0 for that day. Code that assumes a numeric zero will crash or drop the day without saying so. Build any aggregation to treat a missing entry as missing, not as zero engagement.
Put this into practice with AdaptlyPost
Was This Article Helpful?
Let us know what you think!
See us more often in Google
One click marks AdaptlyPost as a preferred source, so our articles sit higher in your Top Stories, AI Mode, and AI Overviews.
Before you go...
AdaptlyPost
Schedule your content across all platforms
Manage all your social media accounts in one place with AdaptlyPost.
All-platform analytics
Social Inbox
AI-powered assistant
Related Glossary Terms


Meta Sets the Instagram API alt_text Character Limit at 1,000
Meta caps the Instagram API alt_text character limit at 1,000 and scopes the field to still images. Reels and stories accept no alt text at all.


Choose Top Analytics Metrics Across the Funnel
Awareness, engagement, and conversion numbers rarely tell one story. A metrics analysis example showing which figures to pair and which to ignore.


A Practical Overview - Audience Analytics Media
Follower counts say almost nothing. The audience analytics media teams rely on cover demographics, behaviour, and engagement, plus where each metric lives.
Related Articles


Turn Instagram Analytics Into Smarter Content
Business and creator accounts get this free. Which Instagram analytics live inside Insights, what each metric means, and where the native tool stops.


Useful Context - Instagram Reach
Counted once per account, Instagram reach is not impressions. What separates organic, paid, and viral reach, and the levers that reliably move the number.


Why the Bluesky Image Size Limit Is 2,000,000 Bytes
The Bluesky image size limit is 2,000,000 bytes per post image, set by maxSize in the images lexicon. Avatars and banners stop at 1,000,000 bytes.

