A view count is a useful signal and a bad witness. Here's exactly what one view is, what it can't distinguish, and how to read the number without inventing a story it doesn't support.
Someone hasn't replied to your proposal. You check the link and it says four views.
What do you now know? Less than it feels like — and more than nothing, which is the interesting part. The view count is genuinely useful and routinely over-read, usually within about ten seconds of looking at it.
It's worth being concrete, because the vagueness is where the over-reading starts.
When someone opens your link, one row gets written: the page it was, and the timestamp. That's the entire record. No name, no email, no device, no location, no idea where they came from.
Two consequences follow immediately, and both cut against how people instinctively read the number.
There's no de-duplication. Every load counts. A refresh counts. Closing the tab and reopening it counts. Coming back tomorrow to re-read the scope section counts. So "four views" spans everything from four different people glancing once to one person who kept returning to the same paragraph — and those are almost opposite situations.
There's no identity. The record has a time and nothing else. Any confidence about who opened it is something you supplied, not something the data did.
That second point deserves its own paragraph, because it's where real mistakes happen. You send a link to one person, see two views, and conclude they read it twice. Perfectly plausible. Also perfectly consistent with them forwarding it to a colleague. The count cannot distinguish those, and the moment you start making decisions as if it can, you're following a story you wrote yourself.
This is the limit worth internalising, because it's structural rather than a missing feature.
An unguessable link is a bearer credential — holding it is the permission, which is exactly what makes it work without an account. The flip side is that a forwarded link opens precisely like an original one. Nothing marks the second reader as second.
So if a document genuinely must not travel, analytics is the wrong instrument. The controls that work are the ones applied before sending: a password, so holding the address isn't sufficient on its own, and an expiry, so the window closes on a date you chose rather than whenever you remember. Watching a number afterwards can't tell you a forward happened, and a control that only tells you after the fact wouldn't be much of a control anyway.
Every page carries a view count, on every plan. Solo and Studio add the last-opened time and a day-by-day trend for the past 30 days.
Having said all that — the number earns its place, as long as you ask it the question it can answer.
It separates two kinds of silence. This is the whole value, and it's larger than it sounds. A client who hasn't replied and hasn't opened it has a full inbox; the right move is a short nudge with the link repeated. A client who hasn't replied and opened it four times has a question they haven't asked; the right move is a call, or an email that pre-empts the objection you suspect. Before you had the number, both looked identical from your side and you picked one at random.
Timing reads better than volume. A total tells you little. A shape tells you a lot — opened within an hour of sending is a different signal from opened on day six, and a cluster of opens on a Tuesday morning usually means it got circulated internally before a meeting. That's why the day-by-day view matters more than the count: three views on one afternoon and three views across three weeks are not the same event.
It tells you when your work went cold. Steady opens that stop is a useful thing to notice about a page you're keeping live for a client.
Most of these counts are small — single digits — and small numbers are exactly where people over-infer.
Some discipline that helps:
That last point is the one real path to identity: putting a short email step in front of the page, so a reader submits an address before it opens. It's a genuine trade rather than a free upgrade — you're adding friction to a page whose main virtue was opening on click, and some readers will simply leave. It suits a lead magnet, where capturing the address is the point. It suits a client proposal badly, because your client should not have to fill in a form to read something they commissioned. Even then, what you learn is who submitted an email, which still isn't quite who read the document.
It would be easy to read all this as a list of gaps. It's closer to the opposite.
A page that fingerprints its readers, records where they came from, and follows them around would produce a richer number and would be a strange thing to send a client — especially from a product whose starting position is that a page you share shouldn't be a surveillance surface. The recipient of your proposal didn't agree to be measured; they agreed to read something. There's a version of "analytics" that quietly forgets this, and the counts it produces aren't actually better for the decision you're making. You still wouldn't know whether they read it. You'd just have more numbers around the same uncertainty.
What you want from this signal is modest and achievable: has this been opened, roughly when, and is the pattern still moving. That's enough to decide whether to follow up today or wait until Thursday — which is the actual decision — and it's the same reason sending a link rather than a file is worth doing at all. A copy that has left your hands can't tell you anything.
Read it as a smoke alarm, not a security camera. It tells you something is or isn't happening. It does not tell you the story, and the moment you ask it to, it starts making one up.
A shared link tells you more than that it was delivered — it tells you when someone opened it. Here is how a view timestamp on a private link turns client, sales, and creator follow-up from guesswork into timing.
The attachment tells you nothing after it leaves. A link tells you it was opened, when, and for how long — which is most of what you wanted and less than people assume. What the number actually means, and where it stops being evidence.
You open your host's protection settings and get an upgrade screen instead. The question isn't how to add a password — it's whose free plan has that switch, what the switch actually locks, and when paying for it beats moving.
miinideck turns a single HTML file into an unguessable link with optional password and expiry. Default-private, never indexed.