How to structure the page is a settled argument: one permanent address, one obvious next action, a short list underneath, pruned often. That case is made in full in the link in bio field guide, it is still right, and this post does not relitigate it.
What follows is the other half — the practices that decide whether a structurally correct page actually works in 2026, on the phone it will really be opened on. Almost none of them are design decisions, and every one of them is fixable this afternoon.
Design for where the tap came from
Nobody reaches your page from nowhere. They arrive from a profile they were already reading, from a story, from an automated DM that promised them one specific thing, or from a QR code on a table. Those are four different states of mind, and a page written for only the first quietly fails the other three.
The profile visitor is browsing and needs orientation. The person who came from a DM was promised something exact and wants only that. If your page makes the promised thing hard to find, the automation you built has delivered somebody to a search problem — so give it its own row, in the words the message used, where a thumb lands first.
From the profile: orientate.
Name, face, and what you do in one sentence a stranger could repeat. They do not know you yet.
From a DM: deliver.
They were promised one thing. It should be the first thing, worded the way the message worded it.
From a QR code: assume no context.
Somebody scanned a card or a screen. Nothing about the room they are standing in is on the page.
Write the first line, not just the links
The most-read words on a bio link page are not the links. They are the line or two above them, and most creators spend that space on a job title. A job title tells a stranger what you are called; it does not tell them whether the next tap is meant for them.
Write it as a sentence you would say out loud — what you make, who it is for, and what is worth their attention right now. It qualifies the visitor inside the four seconds you get, and it is usually the same line that follows your address into a shared message or a search result, read by somebody with no other context.
Say what somebody gets, not what you are.
“Ten-minute mobility routines, three mornings a week” beats “Coach · Creator · London”.
Keep it under two lines on a phone.
The links have to be visible without a scroll. Everything above them is spending that budget.
It doubles as your preview text.
Write it to survive being quoted out of context, because that is exactly what happens when the link is shared.
The in-app browser is the real platform
Almost every visit opens inside Instagram’s or TikTok’s own webview rather than in the browser the person actually uses, and that environment is narrower than the open web in ways worth designing around. Extensions do not exist. Third-party cookies often do not survive. Saved passwords and autofill are frequently unavailable, and the back gesture returns to the feed rather than to you.
The consequences are practical rather than theoretical. Anything asking a visitor to sign in somewhere else converts worse than you expect, because their password manager is not there. Anything depending on a cookie set on another domain may never fire. Test by tapping your own bio link from inside Instagram, on a phone, and fix what you find there rather than what you saw on a laptop.
Tap targets, not click targets.
A row a thumb can hit without aiming, and nothing important pressed against the edge of the screen.
No step that needs a password manager.
If a destination demands a login, expect a share of people to stop there. Offer the version that does not.
Assume your analytics under-count.
Webviews and tracking prevention lose events. Trends are trustworthy; absolute numbers are not.
Give the page a speed budget
Most visits arrive on mobile data from somebody who has already decided to give you a few seconds. A page taking three of them to paint has spent most of that budget before the first link appears, and the person who leaves at that point never shows up as a visit you lost.
The fixes are unglamorous and permanent. Compress the header image instead of uploading the four-megabyte original. Keep one screen of real content above the fold. Resist the auto-playing video embed at the top, which is usually the heaviest thing on the page and is competing with the links you actually want tapped.
One hero image, sized for a phone.
Cropped to the dimensions it renders at. Nobody needs your camera’s full resolution on a six-inch screen.
Embeds below the fold.
A player can cost more to load than everything else combined. Put it under the rows it competes with.
Measure on data, not on wifi.
The real conditions are worse than yours, and they are the only conditions that count.
Make the traffic tell you where it came from
The usual complaint about a bio link page is that it reports taps without explaining them. That is normally fixable without buying anything: give each campaign its own words. A distinct keyword for each automated DM, a distinct label on the row it points at, and a month of activity stops being one undifferentiated number.
One promise, one row, one keyword. When the same phrase appears in the caption, in the automated message and on the row somebody taps, you can tell which post did the work — and you can retire a campaign later without wondering what else was quietly riding on it.
One keyword per promise.
Reusing a keyword across two offers makes both reports meaningless and both messages slightly wrong.
Name the row after the promise.
“The ten-minute reset” is a label and a measurement at once. “Free guide” is neither.
Date what you add.
A row you cannot date is a row you will never confidently archive, so it stays forever.
Keep the page and the DM saying the same words
The page rarely fails on its own. It fails at the seam, where an automated message says one thing and the page it points at says something adjacent. Somebody who commented RESET and lands on a row headed “New program — enrol now” has to translate before they can act, and most people simply do not bother.
So review the two together rather than separately: read the automated message and the page in one sitting, and change whichever one has started lying. If you have not built the automation half, the step-by-step comment-to-DM setup takes about two minutes, and the link in bio for creators page covers what the two halves look like as one product.
The same noun on both sides.
Whatever your audience calls the thing, use that word in the caption, in the message and on the row.
Edit them in one sitting.
A page change that does not check the DM is how the seam opens in the first place.
Retire them together.
An automation still pointing at a row you archived is a dead end only your audience can see.
Frequently asked questions
How many links should a link-in-bio page have?
Three to five live rows, plus one clearly prioritised action at the top. Longer lists reliably lower the tap rate of everything above them, because a page with everything on it is a page where nothing gets chosen.
Should I design for the in-app browser or a real browser?
The in-app one. Most taps open inside the app’s own webview, so design for mobile widths, large tap targets, no browser extensions, and flows that still work when third-party cookies and autofill do not.
Is a link-in-bio page bad for SEO?
It is neutral on its own. A creator page becomes a genuine search asset when it has a stable address, a descriptive title, an accurate description, and it is not the only place your work lives. Treat it as a front door, not an entire site.
How often should I update the page?
Change the top row whenever your content starts promising something new, and read the page and your automated DMs together once a month. The address itself should never change.
Fix the seams, not the theme.
The structure is settled. What is left in 2026 is the unglamorous half — where the tap came from, what the webview permits, how fast the page paints, and whether it still tells the same story as the message that sent somebody there. Ten minutes a month keeps all four honest.
Put it into practice