The link tax: does putting a link in a post cost you reach?

You are writing the caption. The post is ready. You have a link that would help. A source. A tool you mentioned. Your own store. You start to type it and stop. Everyone knows you do not put links in posts. Links get you throttled. Instagram wants to keep people on the app. So the link goes nowhere, or into your bio, or not at all.

That belief is one of the stickiest pieces of platform advice there is. It shows up in every creator forum. It turns up in the replies under every reach complaint. And it makes perfect sense. Instagram makes money when you stay on Instagram. Every click away is a click they do not profit from. So of course they would suppress posts that send people away.

The question is not whether the belief is plausible. It is. The question is whether it is documented. We went looking for the documentation. We read Meta's system cards. Feed, Explore and Reels. We read Adam Mosseri's ranking post. We searched for Instagram engineering posts about links. And we checked what Mosseri has said when asked.

Here is what we found. Instagram has never said external links reduce reach. Not in writing. The system cards list ranking signals. Links are not on those lists. Mosseri's May 2023 post explains ranking. It does not mention links. And when asked about one version of the claim, he denied it.

Facebook is a different story. Facebook has said it suppresses one thing. Clickbait headlines. Posts with misleading or exaggerated titles. Ones that trick people into clicking. That is not about links. That is about lies. And that is Facebook. Not Instagram.

So the trail is short. The incentive argument is sound. The documentation is missing. And the one direct answer Instagram's head has given goes the other way.

Thin reach usually has a duller explanation than a hidden penalty. A new post arrives as a cold item, with no history for the models to read. A link penalty would be a separate mechanism on top of that. It would need its own evidence.

What follows is what we read, where we looked, and where the line between argument and evidence sits. Every claim below is linked, quoted and dated. The sources are at the back.

The system cards do not mention links

Meta publishes a page for each part of Instagram. Each page explains what that part tries to guess about you and what facts it reads to make the guess. Meta calls these pages system cards. They sit on the Meta Transparency Center. A site Meta runs to explain how its software works.

We read four Instagram cards. One for Feed. One for Explore. One for Reels Chaining. Chaining is Meta's word for the run of reels that keeps going once you open one. And one for Feed Recommendations. That last one covers posts in your feed from accounts you do not follow.

Each card lists the guesses its models make. They guess whether you will like a post. Whether you will skip it. Whether you will spend time on the author's profile after seeing it. Whether you will share it in a message. Whether you will ask to see less of it. Then the card lists the facts the models read to make those guesses. Meta calls those facts signals.

Not one of the four cards names external links as a signal. Not as something the models read. Not as something they penalize. Not anywhere.

The Feed card lists fifteen signals. Here is the full list, copied from the card stamped 29 June 2026:

• How likely you are to spend time on a post
• How likely you are to comment on a post
• How likely you are to like a post
• How likely you are to save a post
• How likely you are to tap on a profile photo after seeing a post
• How likely you are to share a post with someone in a direct message
• How much time you're likely to spend on the author's profile after seeing this post
• How likely you are to watch a video and for how long
• How likely the first post you see will keep you engaged with your feed for the rest of your session
• How likely you are to skip the post
• Your past interactions with the author
• The post format (e.g., photo, video, carousel)
• How many cards are in the carousel
• The predicted value of text overlay on the post
• How recently the post was created (within 90 days)

Links are not on that list. Neither is "sends people off Instagram". Neither is "competes with our ad revenue".

The Explore card lists ten signals. The Reels Chaining card lists eight. The Feed Recommendations card lists eleven. Not one of them mentions external links. The only thing any card says about links is under Feed Recommendations. There the card names one guess the model makes: "How likely you are to tap on a link in a post". That counts tapping. It says nothing about the guess being negative.

Adam Mosseri is the head of Instagram. On 31 May 2023 he published Instagram Ranking Explained. It walks through Feed, Stories, Explore and Reels. It names the most important signals. In rough order. For Feed, those are:

Your activity, Information about the post, Information about the person who posted, Your history of interacting with someone — Adam Mosseri, Instagram Ranking Explained, 31 May 2023

The post lists examples under each one. How many times you liked posts with a given filter. How popular a post is with other people. How many times you interacted with the author in recent weeks. It names nothing about external links. Not as a signal. Not as a penalty. Not anywhere in the post.

So the public record has five ranking documents. Four system cards and Mosseri's explanation. Together they list 44 named signals across four parts of the app. Zero mention links.

Mosseri has denied one version of the claim

The belief turns up in different forms. One of the most common is about the phrase "link in bio". The claim: saying "link in bio" in a caption will tank your reach. Instagram sees those words. Then it suppresses the post.

Mosseri addressed that one directly. He did it on Threads. Threads is Meta's text app. The statement was in July 2025. We could not retrieve the original post. But two trade publications quoted it the same day. Both carried the same quote. From Social Media Today, 23 July 2025:

if you say 'link in bio' it's going to decrease your reach. That is not true. You're more than welcome to say 'link in bio,' it will not affect your reach — Adam Mosseri, quoted in Social Media Today, 23 July 2025

The same article reports Mosseri adding "it will not affect your reach one way or another". And that he uses the phrase himself regularly. That is as direct a denial as you can get. The head of Instagram. On the record. Saying a widely believed penalty does not exist.

That is one version of the claim. "Link in bio" is a phrase. It is not the same as an external link in the caption. Or the first comment. Mosseri's statement covers the phrase. It says nothing about actual links. But it does tell you something. When asked directly about a folk belief, he denied it. He did not dodge. He did not say "it depends". He said it is not true.

What Facebook has said is about clickbait, not links

Facebook and Instagram are both Meta products. They share some ranking ideas. But they are not the same app. And their public documents do not line up.

Facebook has published multiple announcements about suppressing one kind of post. Posts with clickbait headlines. On 4 August 2016, Facebook announced it would reduce clickbait in feed. The post defined clickbait. Headlines that fall into two groups:

if the headline withholds information required to understand what the content of the article is; if the headline exaggerates the article to create misleading expectations for the reader — Facebook News Feed FYI, 4 August 2016

The examples Facebook gave were headlines like "When She Looked Under Her Couch Cushions And Saw THIS… I Was SHOCKED!" That kind. "He Put Garlic In His Shoes Before Going To Bed And What Happens Next Is Hard To Believe". Those are not links. Those are lies. Or at least tricks. Facebook said it built a system to identify common clickbait phrases. And it said this:

Links posted from or shared from Pages or domains that consistently post clickbait headlines will appear lower in News Feed. — Facebook News Feed FYI, 4 August 2016

That is about Facebook. Not Instagram. And it is about misleading headlines. Not external links. The two get confused constantly. A clickbait post usually has a link. That is how you click the bait. But what Facebook said it penalizes is the misleading part. Not the link part.

Facebook updated that system in May 2017. The update evaluated clickbait at the post level. Not the domain level. So one clickbait post would not sink a whole site. The announcement repeated the two-part definition. Withholding information. Or exaggerating it. It said nothing about penalizing ordinary external links.

The Meta Transparency Center has a page on clickbait links. That page describes the policy. It is under the Facebook section. Not Instagram. And the policy is the same. It targets "posts that lure people into clicking on an included link by creating misleading expectations". Not posts with links. Posts with lies.

So Facebook documents one thing it demotes. Manipulative headlines. That is not the same as external links. And even if it were, Instagram is not Facebook. The Instagram system cards carry their own signal lists. None of them say "we copied Facebook's clickbait rules".

The incentive argument is real but it is not evidence

Here is why the belief makes sense. Instagram is a business. It makes money by showing you ads. The longer you stay on Instagram, the more ads you see. The more ads you see, the more Instagram earns. Every time you click a link and leave, that chain breaks. You stop seeing ads. So Instagram has a reason to keep you there. And the way to do that is to show you posts that keep you there. Not posts that send you away.

That argument is correct. It describes Instagram's business model. Ads fund the app. Attention is the currency. Keeping you on the platform is worth money. Losing you to an external link costs money. Any ranking system would account for that.

But an incentive is not the same as a policy. And a policy is not the same as a document. Instagram has an incentive to favor native content over link posts. That does not mean it does. And if it does, that does not mean it has said so.

The incentive argument tells you what Instagram might do. It does not tell you what Instagram does do. For that you need a statement. An engineering post. A system card that names the penalty. A Mosseri post that says "yes, we reduce reach for external links because they take people off the app". We went looking for that statement. We could not find it.

What we found instead is silence. The system cards name dozens of signals. Links are not among them. Mosseri has explained ranking many times. He has never said links hurt. When asked about one version of the claim, he denied it. That is the opposite of proof.

The one study that exists is not a study

We found one piece of data on Instagram and external links. It is a blog post by Hootsuite. Published in March 2023. The author looked at their own Instagram account. They checked their top 20 Stories across four measures. Replies, reach, shares and follows. Then they counted how many of those top posts had a URL sticker. A sticker is how you add a link to a Story.

The results:

• 20% of top Stories by replies had links. 80% did not.
• 20% of top Stories by shares had links. 80% did not.
• 25% of top Stories by reach had links. 75% did not.
• 30% of top Stories by follows had links. 70% did not.

Stories with links were less likely to be top performers. That is what the numbers show. But what do those numbers mean?

The sample is one account. The author's own. One person's Stories over a few months. No control group. No test. No check on whether the linked Stories were about different topics than the ones without links. No attempt to hold anything constant except the link.

The author knows why the pattern exists. Links "ask your followers to leave Instagram" instead of engaging on it. That is the same incentive argument. It is not proof Instagram punishes links. It could just as easily be that people do not click away when they are in Story mode. Or that the linked Stories were less compelling to start with. Or that the author posted links on different topics than they posted native content about.

A real test would randomize. Take one piece of content. Post it twice. Once with a link. Once without. Same image. Same caption. Same time of day. Same audience. Then measure the reach. Do that a hundred times. That would separate the effect of the link from the effect of everything else.

This is not that. This is one person noticing a pattern in their own archive. That is observation. It is not evidence of ranking penalties. And it is not a controlled test.

No peer-reviewed study we could find has tested whether Instagram suppresses external links. No published test. No data release from Meta. The Hootsuite post is the closest thing to data that turns up in a search. And it does not prove the claim.

Where the line between documented and assumed sits

Here is what is documented.

Instagram ranks posts by guessing how you will interact with them. It names those guesses in the system cards. It names the signals that feed those guesses. External links are not among the signals named. That is in four system cards. All retrieved 17 August 2026.

Adam Mosseri has denied that saying "link in bio" reduces reach. That is in trade press coverage of a Threads post from July 2025.

Facebook demotes posts with clickbait headlines. Ones that withhold information or exaggerate to mislead. That is in Facebook News Feed posts from 2016 and 2017.

Instagram makes money by keeping you on Instagram. That is how ad-supported apps work.

Here is what is not documented.

No Instagram system card says external links reduce reach. No Mosseri post says it. No Meta engineering blog says it. No published research tests it with a controlled test. The incentive argument is strong. The folklore is everywhere. The evidence is absent.

The belief that links cost you reach drives real behavior. Creators avoid linking out. They send people to their bio instead. They save links for Stories. Or keep them out of captions. Some never link at all. All of that is based on a penalty no one can point to in writing.

We are not saying the penalty does not exist. We are saying it is not documented. Those are different claims. Instagram could suppress external links and never say so. Companies do not publish every detail of how they rank things. Silence is not proof of absence.

But silence is also not proof of presence. When a belief is this widespread and this specific, the burden is on the claim. Show us the document. Show us the engineering post. Show us the system card entry. Show us the test. If none of those exist, what you have is not knowledge. It is assumption.

That assumption may be correct. The incentive is there. The pattern some people see in their own reach is real to them. But an incentive and a pattern are not the same as a published penalty. And treating them as the same thing is how folk beliefs become uncheckable.

What would prove this wrong

Three things would prove this post wrong. And we would update it the day we caught them.

First, Meta publishes a system card update that names external links as a negative ranking signal. If a future version of the Feed card or Explore card says "How likely a post is to send you off Instagram" and lists that as a signal that lowers the score, then everything above about undocumented penalties is wrong. The penalty would be documented.

Second, Mosseri or another Instagram engineer publishes a post saying external links reduce reach. If Mosseri writes "we reduce distribution for posts with external links because they take people off the app", that would be the statement we could not find.

Third, a peer-reviewed study runs a controlled test on Instagram and finds that external links reduce reach. Take the same post. Publish it with and without a link. Randomize which accounts see which version. Measure the reach. If the linked version performs worse and the difference is significant and no other variable explains it, that would be evidence.

None of those exist today. If any of them appear, this post gets a dated correction in the changelog. With the source.

What we could not determine

We could not retrieve Adam Mosseri's original Threads post from July 2025. We read two trade press summaries of it. Published the same day. Quoting the same words. Both sources are in the list at the back. But we did not read the original post. Threads posts are hard to retrieve outside the app. So the Mosseri quote is second-hand. Accurately reported by two independent outlets. But second-hand.

We could not find the exact date of Mosseri's Threads statement. The trade press articles are dated 23 July 2025. Neither names the date Mosseri posted. So we know the statement was recent as of that date. We do not know the day.

We could not load some Meta Transparency Center pages fully in an automated fetch. The pages are built by code that runs in a browser. A plain download returns an empty shell. So every Transparency Center page cited here was opened and read in a browser. Not fetched as a file. That means we read what a person sees. Not what a machine retrieves. Both are legitimate. But they are different. And we are noting the difference.

We could not determine if Instagram uses any Facebook clickbait rules. Facebook's documentation is separate from Instagram's. The two apps share a parent company. They do not share documentation. So we report what Facebook publishes about clickbait. We do not claim Instagram uses the same system. It might. The documents do not say.

We could not find a single peer-reviewed study testing external links on Instagram. If one exists and we missed it, send it. We will read it and update this post.

How we did this

We read published documents. No data collection. No test. No access to Instagram's internal systems.

Every source was opened on 17 August 2026. Every quote was copied from the page as a reader sees it. Meta Transparency Center pages and the Instagram Help Center were read in a browser. Those pages are built by code and return empty shells to a plain fetch. The two Meta engineering posts and Mosseri's ranking post were retrieved as finished pages.

The Mosseri Threads quote was read through two trade publications. Not directly. Threads posts are difficult to retrieve outside the app. Both outlets quoted the same words on the same day. That is reported as second-hand coverage. Not a primary read.

One typographic change is applied to every quotation. Curly apostrophes and quote marks are rendered as straight ones here. No other character is altered. No word is changed.

Limits: twelve of fourteen sources are first-party statements. All from Meta or Instagram. A system card is a description the company chose to publish. Not an audited account. Meta rewrites these cards without saying when. So every card quotation carries the date we read it.

Sources

1. Instagram Feed AI system card, stamped UPDATED JUN 29, 2026. Meta Transparency Center. Read in a browser. Retrieved 17 August 2026.

1. Instagram Explore AI system card, stamped UPDATED JUN 22, 2026. Meta Transparency Center. Read in a browser. Retrieved 17 August 2026.

1. Instagram Reels Chaining AI system card, stamped UPDATED NOV 11, 2025. Meta Transparency Center. Read in a browser. Retrieved 17 August 2026.

1. Instagram Feed Recommendations AI system card, stamped UPDATED JUN 29, 2026. Meta Transparency Center. Read in a browser. Retrieved 17 August 2026.

1. Adam Mosseri, Instagram Ranking Explained, 31 May 2023. About Instagram. Retrieved 17 August 2026.

1. Sara McCorquodale, Instagram doesn't penalize posts that include 'link in bio', Social Media Today, 23 July 2025. Reports Mosseri's Threads statement. We could not retrieve the original Threads post. Retrieved 17 August 2026.

1. Instagram confirms 'link in bio' won't hurt reach, SMK, 24 July 2025. Second source reporting the same Mosseri statement. Retrieved 17 August 2026.

1. Alex Peysakhovich and Kristin Hendrix, News Feed FYI: Further Reducing Clickbait in Feed, 4 August 2016. Facebook. Retrieved 17 August 2026.

1. News Feed FYI: New Updates to Reduce Clickbait Headlines, 17 May 2017. Facebook. Retrieved 17 August 2026.

1. Clickbait links, undated. Meta Transparency Center, listed under Facebook. Read in a browser. Retrieved 17 August 2026.

1. Amanda Wood, Do links in Instagram Stories ruin your engagement?, Hootsuite blog, 20 March 2023. Observational data from one personal account. No sample size stated for the account's total Stories. Top 20 in each metric category analyzed. No control group, no statistical testing. Retrieved 17 August 2026.

Changelog

• 2026-08-18: First version drafted.