How Creator Agencies Track Which Platform Actually Converts
Short answer
Give every creator × traffic source combination its own link, name them by a fixed scheme, and compare three counters: the link page's visits, its outbound clicks, and the destination platform's own counter for that tracking link. The gap between the last two is where in-app browsers eat your traffic.
Every agency hits the same wall around the third creator. The traffic is real, the revenue is real, and nobody can say which network produced it — so the posting schedule is decided by feel. The fix is not a better dashboard. It is deciding what one link is allowed to mean, before you create three hundred of them.
Why can't you tell which platform converts?
Because of a shortcut that looks efficient at the start. One creator gets one link, that link goes in every bio, and the numbers add up into a single figure. It scales beautifully and it destroys the only question worth asking.
A blended number tells you traffic arrived. It cannot tell you that one network sends ten times the volume at a tenth of the conversion rate, which is the single most common shape in this business and the one that decides where the next hour of work goes. Worse, the blend is stable: it barely moves when one source collapses, so a dead network stays in the rotation for months.
What should one link represent?
One creator on one traffic source. Nothing more, nothing less.
That is the smallest unit you can actually act on. You can redirect it, pause it or replace its destination without touching the other eleven. Its numbers mean one thing, so a change in them has one possible cause. And when a creator leaves or a network dies, you delete a row rather than untangling a shared counter.
The cost is volume: five creators across four networks is twenty links, not five. That is the correct trade. Twenty links you can read beat five you cannot.
What naming scheme should you use?
Any scheme works as long as it is mechanical, so pick one and never deviate. The shape that survives contact with a real roster is creator, then traffic source, with no separators and consistent casing:
AmiraHadiXTelegram
AmiraHadiXReddit
LenaKochXTikTokThree properties earn their keep. It sorts by creator, so a roster of sixty is still readable. It is derivable — given a creator and a source, there is exactly one correct name, so two people building links on different days build the same ones. And it is greppable: you can find every link for one creator, or every link on one source, with a substring.
Two failure modes to design out from day one. Spelling drift is the expensive one: X, Twitter, x.com and X (Twitter) are one network and four different rows, and once they exist every filter silently lies to you. Fix the vocabulary once, write it down, and fold every alias onto the canonical spelling before a link is created — not in a cleanup pass six months later. The same goes for creator names with accents or casing variants.
The second is overloading the source field with the destination. Keep them separate: the traffic source is where the click came from, and the destination is where it goes. Conflating them is why so many agency spreadsheets cannot answer either question.
Which numbers actually exist, and what does each one measure?
Three counters, measuring three different things. Confusing them is where most agency reporting goes wrong.
Swipe the table sideways to see every column →
| Counter | What it counts | What it cannot tell you |
|---|---|---|
| Page visits on your link | The page opened | Whether anything was clicked afterwards |
| Outbound clicks on your link | A visitor tapped through to the destination | Whether they arrived |
| The destination's own counter | Arrivals the platform recognised | Where those arrivals came from |
The first two you own. The third belongs to whichever platform you send people to, and most of them expose some per-link counter for exactly this purpose.
The interesting figure is not any of the three. It is the ratio of the third to the second — how many of the people who tapped through actually landed. That is your pass-through rate, and it is the only number in the set that describes a leak rather than a volume.
Why does a pass-through rate above 100% appear?
Because two numbers covering different time spans got divided by each other. This is worth spelling out, because we shipped it ourselves and it produced readings of both 818% and 1% on the same screen.
If your outbound clicks cover the last 30 days and the destination's counter has been accumulating since the link was created, the ratio is arithmetic nonsense — too high for an old link, absurdly low for a new one, and never comparable between two rows. A number that moves for reasons unrelated to reality is worse than no number, because people act on it.
Two rules fix it permanently. Compare like-for-like periods only: if the destination only exposes a lifetime total, then the only honest comparison is your lifetime total against it. And snapshot on a fixed day — write both numbers down weekly, on the same weekday, so the difference between consecutive snapshots gives you a real windowed figure on both sides.
How do you join your numbers to the platform's?
On the destination URL, which is the one value both sides know. It needs normalising before it will match, because the same address has several legitimate spellings: lowercase the host, drop a leading www., strip the trailing slash, and remove the query string and fragment before comparing.
Two operational rules make this stay correct. Keep the raw URL as well as the normalised one, so nothing is lost if the rule changes. And never drop a destination you cannot match — show it separately as "traffic with no matching link". An unmatched row is usually a renamed or re-created tracking link, and hiding it is how a whole creator quietly disappears from a report.
What is the drop-off actually made of?
Once you can see the gap, the next question is what fills it. Three things, in roughly descending order:
- In-app browser failures. The click happened inside Instagram, TikTok, Threads or X, in a webview with an isolated cookie store and a blocked app hand-off — so the visitor arrived logged out, hit a sign-in or verification wall, and left. This is the largest share for most agencies and the only one you can fix from your side.
- Interstitial bounce. Someone tapped, saw the intermediate page, and changed their mind. Real, and mostly a design question.
- Geo and age restrictions on the destination. Outside your control, but stable over time — so it shows up as a constant floor rather than a change.
The reason the first one dominates is that it is the default case, not the edge case: the overwhelming majority of taps on a creator link happen inside one of these apps. We broke down exactly which capabilities disappear in in-app browser vs real browser.
How do you run this with no tooling at all?
You can have the entire method today, in a spreadsheet, for nothing. It is worth doing even if you later automate it, because it forces the vocabulary decisions that automation will otherwise bake in wrong.
- One row per creator × traffic source. Columns: creator, source, link name, destination URL, and one column per counter.
- One tracking link per row, created on the destination platform, named by the scheme above.
- A fixed snapshot day. Every Monday, write down both counters. The difference between this week and last week is your real weekly figure on both sides.
- A canonical vocabulary tab — the exact spelling of every network and every creator name. Everything gets typed from that tab, never freehand.
- An unmatched bucket for destinations that do not line up with a row, reviewed weekly rather than deleted.
That is the whole system. Most agencies never get further than this and most do not need to.
What does plugwith.me remove?
The manual half, and the largest line item in the drop-off.
Each link is its own page with its own visit and click data, filed under a creator and a traffic source, so the spreadsheet's two hardest columns fill themselves and the filters use one canonical spelling instead of four. Creating a link for a new creator × source pair is one form rather than a copy-paste ritual, which is what actually keeps a naming scheme intact across a team.
More importantly, every one of those links escapes the in-app browser: when a visitor taps from inside Instagram, TikTok, Threads or X, the page hands them to their real browser first, so the destination sees a normal visit with a real session instead of a logged-out stranger. That converts the biggest component of your pass-through gap into arrivals. The mechanism, per platform and with its honest limits — no scheme escapes Facebook or Messenger today — is documented in the pillar guide.
Before you change anything, measure the thing you are about to fix: open the free WebView test from one of your own bios, on a real phone. It reports which environment the tap actually happens in and what is unavailable there, which is usually the moment the gap in the spreadsheet stops being abstract.
Frequently asked questions
Why is one link per creator not enough?
Because it produces one blended number. If a creator posts on four networks and all four point at the same link, you can see that traffic arrived but not which network sent the people who actually converted, so you cannot decide where to spend the next hour.
What should one link represent?
Exactly one creator on exactly one traffic source. That is the smallest unit you can act on: you can change, pause or redirect it without touching anything else, and its numbers mean one thing.
Why does my pass-through percentage come out above 100%?
Almost always because a windowed number is being divided by a lifetime one. If your link clicks cover the last 30 days and the destination's counter has been running since the link was created, the ratio is arithmetic nonsense in both directions.
Do I need software to run this?
No. A spreadsheet with one row per creator × source, a fixed weekly snapshot day and one link per row gives you the whole method. Tooling removes the manual snapshot, not the thinking.
What does the gap between link clicks and platform clicks actually contain?
Three things: visitors who bounced on the interstitial, visitors blocked by geo or age restrictions on the destination, and visitors whose hand-off failed inside an in-app browser. The third is usually the largest and the only one you can fix from your side.
Make every tap open the real app.
Built for Agencies · Attribution — and every other platform. One plugwith.me link escapes the in-app browser and hands off to the native app — automatically.
Create your free link →