The confusion is not really about clicks. It is about the fact that two reasonable people can define "unique" differently and both be right, because uniqueness is measured against an identifier and a time window that the tool chooses, not a universal standard.
What a unique click actually counts
A unique click credits one click to a visitor identified by whatever the tool uses, a cookie, an IP and user agent hash, or a logged-in ID, and only counts that visitor once inside a defined window. Click again five minutes later from the same browser inside that window, and the unique count does not move. Click again tomorrow, or from a different device, and depending on the window it might count as a second unique. Some tools set that window to a single session, expiring the moment the browser closes, while others persist across days or weeks using a longer-lived cookie, and the choice changes how often a returning visitor gets counted as new.
What total clicks counts
Total clicks has no memory. Every click event increments it, whether that is the same person clicking three times while comparing prices, a page reload that re-fires the click event, or a bot hitting the link repeatedly. This is why total clicks is always equal to or greater than unique clicks on the same link, and why a large gap between the two is not automatically bad. It usually just means people are clicking more than once, which is common on pricing links, forms, and anything someone might open, close and reopen while deciding.
Why "unique" is a policy, not a fixed rule
This is the part most explanations skip. "Unique" is not a raw fact about the traffic, it is a definition the tool applies, and the definition has two moving parts: which identifier counts as "the same person," and how long the counting window stays open before someone is treated as a new unique again.
| Identifier used | What it actually dedupes | Where it breaks down |
|---|---|---|
| First-party cookie | The same browser or device across clicks inside the window | Cleared cookies, private browsing, or a new browser reset it, splitting one person into several uniques |
| IP plus user agent hash | The same network and browser combination | A shared network, like an office or a cafe, can merge several distinct people into one unique |
| Device fingerprinting | A device's signature across sessions | More invasive than Raydar or most link tools use, and still resettable |
| Logged-in user ID | The same authenticated account | Only works after login exists, so it is blind to anyone clicking before they sign up |
Window length matters just as much as the identifier. A tool counting uniques over a 24 hour window and a tool counting uniques over the link's entire lifetime will produce different numbers from identical click data, because the shorter window lets the same visitor register as a fresh unique more often. Two tools' unique click counts are only comparable when both the identifier and the window match, and in practice they rarely do. Treat "unique clicks" as a number that is internally consistent within one tool, not one you can safely compare across two.
Click vs visit: two different measurement layers
A click is measured at the redirect, the moment someone activates the link, before their browser has necessarily finished loading anything. A visit or session is measured on the destination page itself, usually by a script that has to load and run first. Those are two separate measurement points, and they can diverge: someone clicks, the click is logged, and then their connection drops, an ad blocker stops the destination page's script, or the page itself fails to load. When clicks run meaningfully higher than visits on the same traffic, that gap is diagnostic. It points at a slow landing page, a broken redirect, or bot activity clicking without ever loading the destination, not just a difference in vocabulary. Raydar's click analytics count at that redirect layer, so referrer, UTM and click ID data stay intact regardless of what happens on the destination page afterward. A click total inflated by automated repeat hits is a different problem than a healthy unique-to-total ratio, and it is worth checking against what bot traffic looks like and how to tell if clicks are real before trusting either number at face value.
How the total-to-unique ratio changes with reach
For a link mostly seen by a small, returning audience, such as one a creator posts in every caption pointing at the same offer, the ratio between total and unique clicks often sits high, because the same core audience clicks again on every new post. For a link that goes wide, shared into a new audience through a repost or an ad, the ratio typically compresses toward close to one to one, because most of that traffic is someone's first and only click. Neither ratio is good or bad by itself. It is far more useful as a comparison against a link's own history than as a fixed target, since a link that suddenly shows a much wider gap between total and unique than it usually does is a signal to check for either genuine renewed interest or repeat automated activity, not a number to chase upward on its own.
A worked example
A link gets five hundred total clicks over a week. Using a thirty day cookie window, the tool logs three hundred and ten unique clicks, which means roughly a hundred and ninety of those clicks were repeats from someone who had already clicked once inside the window, not new interest. Destination page tracking shows only two hundred and seventy five sessions actually loaded afterward, a gap of thirty five against the unique click count. That gap is worth chasing before the click numbers are: it might mean a slow-loading landing page cost you thirty five real visitors, or it might mean thirty five of those unique clicks were automated and never intended to load the page at all. The total click number alone would have hidden both possibilities behind one reassuring headline figure.
The same logic holds for QR codes, which are really just a physical entry point into the same click layer. A poster's code can rack up total scans from the same person testing it twice, but what actually counts as a qr code scan follows the identical unique-versus-total split, cookie or device permitting, since a phone camera opening a link is not fundamentally different from a browser clicking one.
Which number to use for which decision
Use total clicks to gauge engagement momentum on a single post or link, since repeat interest still counts toward how much attention it's getting. Use unique clicks to size your actual audience, especially when you're reporting reach or comparing two links with very different repeat-click behaviour. And use the click-to-visit gap to check whether the destination page itself is loading properly, which is a separate question from whether the link got clicked at all.
Picture two Instagram posts linking to the same product. Post A gets eight hundred total clicks from five hundred unique visitors. Post B gets five hundred total clicks from four hundred and eighty unique visitors. Total clicks alone favours Post A, but Post B reached almost as many distinct people with far less repeat clicking, which usually means its caption did a better job convincing someone on the fence the first time, since heavy repeat clicking on Post A often just means people going back and forth before deciding.
None of these numbers replace the others. A link with high total clicks, modest unique clicks and a tight click-to-visit gap is a small audience clicking repeatedly on a page that loads fine, which reads very differently from a link with the same total clicks but a wide gap between clicks and visits, which points at a landing page problem worth fixing before you spend more on driving clicks to it. See what click tracking is and how to track clicks on a link for the setup that produces both numbers in the first place.