The Pinterest Tag: What It Is and How to Install It Properly

If you’ve ever looked at Pinterest analytics and thought “this tells me people clicked, but not whether any of it was worth anything” — the Pinterest tag is the missing half. It’s a small piece of code on your website that reports back what happened *after* someone left Pinterest, which turns “we got 400 clicks” into “we got 400 clicks and eleven of them bought something”.

It has a reputation for being fiddly, and honestly some of that reputation is earned. But the confusion is mostly about order and vocabulary rather than difficulty. Let me walk you through it the way I wish someone had walked me through it.

What the tag actually does

Pinterest can see everything that happens on Pinterest — impressions, saves, clicks on your pin. The moment someone lands on your website, Pinterest goes blind. It has no idea whether they bounced immediately, read for ten minutes, signed up, or spent two hundred pounds.

The tag closes that gap. It sits on your site and, when something meaningful happens, sends a small message back: someone viewed a page, added to cart, signed up, checked out. Pinterest matches that against the people it sent you and can finally tell you which pins produced results rather than just traffic.

That distinction matters more than it sounds. Without it you optimise for clicks, and clicks are a terrible goal — they reward pins that get people to leave Pinterest, not pins that get people to do something useful. Plenty of high-click pins send nothing but bouncers. You cannot see that until you measure the far side.

Install it in this order

1. Create the tag

In your Pinterest business account, go to Ads and then Conversions. Create one tag per website — not one per campaign, not one per product. You’ll get a tag ID, a short number that identifies your site. Everything else hangs off that.

2. Put the base code on every page

The base code goes on every single page of your site, ideally in the head. Every page, including ones that feel irrelevant — the base code is what lets Pinterest recognise a returning visitor later, so gaps in coverage produce gaps in attribution that are almost impossible to diagnose afterwards.

If you’re on a common platform this is usually a field in your theme settings or a tag-manager container rather than editing code by hand. Whatever route you take, the goal is the same: on every page, once, before the interesting stuff happens.

3. Add event codes where things happen

The base code alone says “someone visited”. Event codes say “someone did the thing you care about”. These go on the specific pages or actions that represent success for you.

Don’t install all of them on day one. Pick the single event that represents real value — a completed checkout, a signup, a booking — get it working properly, and confirm you can see it in Pinterest. Then add the others. Installing six events at once and finding none of them work is a genuinely miserable debugging session, and it’s entirely avoidable.

4. Verify it before you trust it

This is the step everyone skips and it’s the reason for most “the tag doesn’t work” complaints. Go through the action yourself — buy something, sign up, whatever — and confirm the event appears. It can take a little while to show up in reporting, so give it some patience, but do not build decisions on a tag you have never watched fire.

When the tag reports nothing

Four questions, in this order. Nearly every problem is one of them, and going in order saves you from the classic mistake of rewriting working code:

Does it fire at all? Open your browser’s network panel, do the action, and look for the request. If nothing goes out, the code isn’t on the page — a caching layer, a theme update or a tag manager that never published are the usual suspects.

Does it fire on the right page? A checkout event belongs on the confirmation page, not the checkout page. Put it on the checkout page and you’ll record a conversion every time someone starts paying and abandons — inflating your numbers in a way that looks like success and is actively misleading.

Does it fire once? Double-firing is common when a tag is installed both directly and through a tag manager. Every conversion counts twice, every number looks great, and you make decisions on fiction. Check for duplicates before you celebrate a suspiciously good week.

And finally: is it being blocked? Ad blockers and browser privacy features stop a meaningful share of browser-side events from ever reaching Pinterest. This isn’t a bug you can fix in your code — it’s a structural limit of browser tracking, and it’s the reason server-side tracking exists.

How long before it means anything

Set expectations properly or you will panic in week one. Pinterest converts slowly compared to most channels, because people save things and come back to them days or weeks later. A tag installed on Monday will look disappointing on Friday, and that tells you nothing at all about whether it works.

Give it a few weeks of normal traffic before you draw any conclusion, and resist the urge to change three things at once while you wait. If you rewrite your pins, adjust the events and switch attribution settings simultaneously, you will never know which change produced the result — and unpicking that afterwards is far harder than being patient in the first place.

The tag is not your only record

One thing worth saying plainly: the Pinterest tag reports to Pinterest, and Pinterest has an understandable interest in the numbers looking good. That doesn’t make it dishonest — its attribution rules are published and defensible — but it does mean you shouldn’t make it your only source of truth.

Keep an independent record too, through your own analytics and properly tagged links. When the two disagree — and they will — you’ll have something to reason with instead of a single number to take on faith. There’s a whole post on why those two numbers never match, and it’s worth reading before the disagreement surprises you.

What to do this week

1. Create the tag and get the base code onto every page.
2. Add exactly one event — the one that represents real value to you.
3. Do the action yourself and confirm it appears in Pinterest.
4. Only then add the remaining events, one at a time.
5. Tag your outbound links so you have an independent record too.

Once that’s running you finally get to ask the question that actually matters: not which pins got clicks, but which pins produced outcomes. That’s a different list from the one you’d guess, and knowing which of your topics genuinely convert — rather than which merely attract — is what makes the rest of your Pinterest work compound. Understanding which keywords and pins drive that is exactly what PinPinterest is built to surface.

Next: claim your website if that’s still outstanding, then work through the full tracking guide.