I spotted a link in a brochure last month and it set off a slightly uncomfortable train of thought. You can't recall a PDF.
The Bit I Noticed
I was reading through a document before it went out. Nothing dramatic, just the usual pass for typos and whether the numbers add up.
There was a link in it. A perfectly sensible one, pointing at a guidance document on a government body's website. Exactly the sort of link you want: authoritative, relevant, and it saves you writing three paragraphs explaining something that somebody else has already explained properly.
Then I thought: what happens when they move that page?
Because they will. Organisations restructure their websites every few years. Departments merge. Someone decides the guidance needs a refresh and publishes it at a new address. None of that is unusual and none of it is anybody's fault. It's just what happens to URLs over a long enough stretch of time.
And the document with the link in it? That's going out as a PDF. It'll get emailed, forwarded, saved to people's desktops, attached to proposals, printed. There is no version of the future where you get all those copies back.
Print People Already Know This
I spent about twenty years in production and marketing. If you've worked anywhere near print, you have this drilled into you early, usually by getting it wrong once.
You sign off artwork. It goes to the printer. Five thousand copies come back on a pallet. If there's a typo on page four, it's on five thousand copies of page four, and the only fix involves a skip.
So print gets checked properly. There's a whole ritual around it, and everyone respects the ritual, because the consequence of not respecting it is sitting there on a pallet looking at you.
PDFs don't get that treatment. They feel digital, and digital feels changeable. You can always update the file, can't you?
You can update your copy. You cannot update the one someone downloaded in March and attached to a tender in November.
A distributed PDF is print. It just doesn't look like it, so it doesn't get the ritual.
Why This One Is Sneaky
The reason I think this is worth writing about is that it doesn't fail loudly.
A broken image fails loudly. Somebody opens the document, sees a grey box, and sends you a message. A broken link doesn't do that. The person clicks it, gets a 404, assumes they've done something wrong or that the thing has just moved, and gets on with their day. They don't tell you. You never find out.
So the failure is silent, delayed, and lands on your credibility rather than theirs. Somebody reading your brochure in eighteen months hits a dead link and quietly downgrades their opinion of how carefully you put things together. Which is a bit unfair, because you did put it together carefully. You just linked to something that moved.
I should be honest here: I've absolutely done this. Not once. The reason I spotted it this time is that I've been on the wrong end of it before, not because I'm especially clever about links.
What I'd Actually Do About It
Three things, none of them difficult.
Don't link to somebody else's page from anything you distribute. Download the document, host your own copy, and link to that. It doesn't have to be a public part of your site. It just has to be an address you control. If the source updates, you replace your copy and the link keeps working.
There's a caveat: check you're allowed to. Most public guidance is fine to rehost, plenty of other things aren't. Worth thirty seconds of checking rather than swapping a broken-link problem for a copyright one.
Keep a register of what you've embedded and where. This is the boring one, and it's the one that does the work. A list: this document contains these links, it was issued on this date, to roughly these people. When the underlying source changes, you can answer "what have we got out there that's now wrong?" in about a minute instead of guessing.
Nobody wants to maintain this. It's a spreadsheet. But the alternative is having no idea what's circulating with your name on it.
Treat it as a design question, not a proofreading one. "What happens to this link in three years" belongs in the conversation where you decide what goes in the document, not in the final check. By the final check you're looking for typos, and this doesn't look like a typo. It looks correct. That's the problem with it.
Where It Doesn't Matter
I don't want to make this into more than it is.
If you're firing off a link in an email, who cares. If the document has a shelf life of a fortnight, who cares. The register is overkill for most things most people produce, and I'd rather you didn't build a bureaucracy around a brochure.
It matters when three things line up: the document goes out to a lot of people, it hangs around for a long time, and the link points at something you don't own. Brochures, sales packs, onboarding documents, policies, anything with "guide" in the title. That's when it's worth the thirty seconds.
The Actual Point
This is a small thing. I'm aware it's a small thing.
But most of the operational problems I get called in to look at aren't dramatic failures. They're small things that were completely reasonable at the time, repeated a few hundred times, discovered eighteen months later by someone who wasn't there when the decision was made.
A link in a brochure is about as small as it gets. It still ends up in a folder on somebody's laptop, being forwarded to a prospect, pointing at nothing.
The question I keep coming back to, on this and most other things: once this leaves the building, what can I still change, and what have I just committed to forever? For a PDF, the honest answer is that you've committed to all of it.
Got a process that quietly commits you to things nobody's tracking? That's the sort of thing our business process automation work tends to turn up. Give me a shout if you want a second pair of eyes on it.