Most articles about a product's analytics are a tour of the dashboard. This one is mostly about why Pinterest analytics are hard — because the constraints that shape our reports are Pinterest's API constraints, they bind every tool on the market, and they explain most of the strange numbers you've seen.
Three of them decide almost everything: a 90-day hard cap on the range you can request, 60 analytics requests per minute, and lifetime metrics only for pins created on or after 20 March 2023. Every vendor lives inside those. Anyone promising otherwise is describing an API that doesn't exist.
So: what the four numbers actually mean, why zeros lie, how we find dead pins without deleting anything you'd miss, and the honest note that Pinterest's own free analytics are enough for most people.
Four metrics matter: impressions, saves, pin clicks (someone opened the pin) and outbound clicks (someone reached your site). Only the last one is traffic.
A zero isn't automatically a fact. It can mean the pin is newer than the report window, or that a rate-limited request failed and got recorded as nothing.
We never auto-delete. Sync → review → queue → confirm. And a pin whose metrics we couldn't verify is never marked safe to delete.
Pinterest's own analytics are free and genuinely good. If you run one account, use them.
The four numbers that matter
Pinterest exposes several organic metrics per pin. Four carry almost all the meaning.
| Metric | What it actually means | What it tells you |
|---|---|---|
| Impressions | Your pin appeared on a screen | Pinterest is distributing it. Says nothing about whether anyone cared. |
| Saves | Someone saved it to their own board | The strongest signal of intent — they're keeping it for later. |
| Pin clicks | Someone opened the pin itself, on Pinterest | Interest. They have not visited your site. |
| Outbound clicks | Someone tapped through to your URL | Traffic. This is the one that becomes a customer. |
Pinterest also returns profile visits, follows, reactions and comments on some responses. They're interesting; they're not the job.
Pin clicks and outbound clicks are not the same number. One is someone looking closer at your pin. The other is someone standing on your website.
Tools and blogs routinely collapse both into "clicks," which destroys the most useful diagnostic Pinterest gives you: the gap between them.
That gap deserves a moment. If a pin earns lots of pin clicks and few outbound clicks, the image did its job — it stopped the scroll and someone opened it — and then the promise failed. They looked, and didn't want it enough to leave Pinterest. That's a copy problem, not a design problem, and it's invisible if you're tracking "clicks."
CTR: the number that isn't one number
People ask what a good Pinterest CTR is. Two problems, and the first is definitional.
There isn't one CTR. Impressions → pin clicks and impressions → outbound clicks are different rates measuring different behaviours. Anyone quoting a benchmark has usually not said which they mean — and the two can move in opposite directions on the same pin.
And we won't give you a benchmark number. There's no published Pinterest figure we can point you to, and the numbers circulating have no methodology attached. It's the same folklore problem as the "45–60 day" seasonal rule: repeated so often it sounds like data.
We could aggregate our customers' pins and publish an "average CTR." It would be a number generated from our own users, with our own selection bias, using whichever definition of CTR flattered us.
That's marketing dressed as research. We refuse competitors' unsourced statistics — including one that supported our own argument — so we're not going to manufacture our own.
The comparison that works is you against you: same account, same niche, same audience. Which of your pins converts impressions into outbound clicks better than your others? That's answerable from data you own, it's free of everyone's selection bias, and it directly tells you what to make more of. An industry average can't do any of that.
Why Pinterest analytics are hard
This is the section that isn't about us, and it's the most useful thing here. Three API facts shape every Pinterest analytics product on the market — including the ones that compete with us.
| The constraint | What it means in practice |
|---|---|
| 90-day range cap | The per-pin analytics endpoint rejects ranges longer than 90 days, and won't start more than 90 days back. You cannot chunk your way to two years of history — the older windows are simply refused. |
| 60 requests per minute | Pinterest's published rate limits put analytics in org_analytics — 60/minute, described as "Fetching user-related analytics data." One call per pin means a 300-pin account takes five minutes of perfectly paced requesting. |
| Lifetime only from 20 March 2023 | Lifetime metrics come from a different call — the pin object with metrics attached — and are available for pin formats created on or after that date. Older standard pins don't expose it. |
These are Pinterest's rules, not any vendor's design choices. The rate limits are published; the 90-day cap we know first-hand because our integration hit the errors. They're the reason our analytics behave the way they do — and everyone else's too.
The 16× fork in the road
Pinterest's rate-limit documentation sorts endpoints into categories, and two of them decide how any analytics tool gets built:
| Category | Limit | What it covers |
|---|---|---|
org_analytics | 60/min | "Fetching user-related analytics data" — the date-range analytics endpoints |
org_read | 1,000/min | Reading objects — including fetching a pin with its metrics attached |
trends_read | 60/min | Pinterest Trends data |
Quoted from Pinterest's published rate limits. Note the gap between the first two rows.
That's roughly 16× more headroom on the pin-object route than the analytics route — and the pin-object route is also the only one that returns lifetime numbers. So the endpoint that's slower to call is also the one that can't see past 90 days.
Which means a tool that reaches for the obvious endpoint (the one with "analytics" in the name) gets both the tighter rate limit and the shorter history, then either crawls or starts throwing zeros. It's not incompetence; it's a fork you can't see until you've walked into it. We walked into it.
Read the first two together, because their interaction is the whole story. The date-range endpoint can't reach past 90 days and it's the rate-limited one. So the only route to genuine lifetime numbers is the pin-object call — which happens to sit in a category with roughly 16× more headroom (1,000 per minute versus 60).
That's why a well-built tool prefers the pin-object route, and why one that hammers the analytics endpoint gets throttled and shows you zeros. It isn't sloppiness so much as a fork in the road that isn't obvious until you've hit it.
If a tool promises complete lifetime analytics on a pin you published in 2021, ask how. The API everyone shares doesn't offer it. Either they're showing you a 90-day window and calling it lifetime, or they've accumulated data since you connected — which is legitimate, but it isn't history.
We do the accumulation thing as a fallback, and we'd rather tell you that than let you assume we can see 2021.
Why your pins show zero
Three reasons, and only one of them means the pin failed. This distinction matters because a zero is a decision input — people delete pins over it.
1. The pin is newer than the window
If a report covers the last 90 days and your pin is 10 days old, the window starts 80 days before the pin existed. Pinterest returns no data for that period — and a naive tool renders "0."
We know this one intimately because we shipped it as a bug. Newly created pins showed zero across the board and looked dead. The fix was to clamp each pin's window to its own creation date rather than the report's start. Obvious in hindsight; invisible until you notice new pins are suspiciously useless.
2. A rate-limited request failed quietly
At 60 requests per minute, a tool fetching 300 pins at once gets throttled. If its error handling turns a failed request into an empty result, that empty result becomes 0 impressions — and looks exactly like a genuinely dead pin.
We shipped this one too: throttled calls were being swallowed and cached as zeros for half an hour. Now failed fetches are retried with pacing, and — crucially — a pin whose metrics we couldn't fetch is flagged as unverified rather than shown as zero.
3. The pin genuinely got nothing
The boring, real one. It happens, and it's fine — most pins are unremarkable. But you can't identify this case until you've ruled out the first two, which is precisely what most dashboards never let you do.
A zero in a Pinterest tool is not automatically a fact. It can be an artefact of how the data was fetched — a window that predates the pin, or a request that quietly failed.
Before you act on a screen of zeros, ask what would look different if the fetch had gone wrong. If the answer is "nothing," you're reading the tool, not the pins.
Top pins
Your best pins, ranked by impressions, saves or outbound clicks — and this data is comparatively reliable, because it comes from account-level endpoints that return every metric in a handful of calls rather than one call per pin. No rate-limit gauntlet.
What top pins are genuinely for isn't congratulation. It's the only honest answer to "what should I make next?" — because it's evidence from your own audience instead of a guess.
The question to ask of a winner isn't "how do I make more of this pin?" It's "why did this one work?" Was it the angle, the season, the search it happened to match, the colour? Then make a genuinely different pin that shares that quality — not a copy.
The most-saved pin from last October is probably a seasonal artefact. Republishing it in March recreates the pin, not the conditions.
Ask what made it work before you decide what to make more of — see seasonal Pinterest pins for why timing explains more winners than design does.
Underperforming pins — and the safety rules
The feature that finds pins which did nothing. It's also a feature that deletes things permanently, so it's worth showing you exactly how conservative the rules are.
What counts as "safe to select"
All of these, together — not any of them:
That last rule is the one we'd point at if you only checked one thing about a cleanup tool. A zero from a failed request and a zero from a failed pin are indistinguishable — so any tool that treats them the same will eventually delete something that was working. Pins with unverifiable metrics get flagged for manual review instead of quietly joining the pile.
Nothing is deleted without two decisions
| Step | What happens | Reversible? | |
|---|---|---|---|
| 1 | Sync | Pins mirrored into a database with metrics | Yes — nothing touched |
| 2 | Review | You look at what's flagged, numbers visible | Yes |
| 3 | Queue | You choose what to line up | Yes — removable |
| 4 | Confirm | A second explicit confirmation | No — Pinterest deletes are permanent |
There's no one-click purge and we won't build one. Deleting pins based on metrics fetched over a rate-limited API is exactly where "convenient" becomes "destroyed something that was working."
The honest part: you probably shouldn't delete them
We built this feature and we'll still tell you it's usually the wrong reflex.
Pinterest is a search engine. A pin that did nothing for six months costs you nothing to keep — it isn't crowding anything out, and there's no published Pinterest statement that dead pins harm your account. We're not going to invent one to justify a tab.
Where cleanup earns its keep is narrower: pins pointing at pages that no longer exist. A sold-out product, a deleted listing, a 404 — those actively waste the clicks they earn and hand someone a bad experience. That's worth fixing, and often the fix is the link, not the pin.
If you're deleting low performers because you read that they hurt you, close the tab and go make new pins. That's the activity with upside.
What to actually do with the numbers
Analytics are for changing what you make, not for feeling informed. Three diagnoses cover most of it.
| The pattern | What it means | The fix |
|---|---|---|
| Impressions up, saves + clicks flat | Pinterest is showing it; nobody wants it | Change the hook — image or promise |
| Saves up, outbound clicks flat | They like the idea, won't leave for it | Fix the title, not the design |
| Outbound clicks up, nothing on your site | The pin over-promised, or the page under-delivers | The problem is off Pinterest |
| Everything flat | Usually too few pins, not bad pins | Make more, in more angles |
Notice that none of these is fixed by posting the same thing more often. If a number doesn't change a decision, it's decoration.
The fourth row deserves more than a table cell, because it's the most common situation and the least welcome diagnosis. When everything is flat, the problem is usually volume, not quality. Pinterest is a search engine: a pin can only be found for searches it's aimed at, so twelve pins are twelve chances. People respond by rewriting the same twelve pins, which is optimising a sample size of twelve. The fix is more angles at more searches — and if you're publishing four pins a month, no dashboard on earth has anything useful to tell you yet.
There's one number worth watching that isn't in any dashboard: the gap between Pinterest's outbound clicks and your site's sessions. Pinterest counts a tap; your analytics count a page load. People tap and leave before the page renders, so outbound clicks are always higher. A small gap is normal. An enormous one usually means your landing page is slow — which is a genuinely useful thing to learn from two tools disagreeing with each other.
When Pinterest's free analytics are enough
Often. This is uncomfortable to publish and it's true.
Pinterest's own analytics are enough if
- You run one Pinterest account
- You want to know what's working — they're the source of truth, always current
- You're not rate-limited against yourself: it's Pinterest's own data, first-party
- You have fewer than a few hundred pins and can look through them
- You want free forever — we have no permanent free tier
A tool starts earning its place when
- You run several Pinterest accounts and need one view
- You want metrics sitting next to the pins you generated, in one workflow
- You need to find pins pointing at dead pages, at scale
- You're asking "which of my 400 pins should I stop making more of?" without opening 400 tabs
- You're an agency and the reporting is the deliverable
If you landed in the left column, use Pinterest's own dashboard and spend the money on nothing. It's free, it's first-party, and it will always be more current than any tool syncing over a rate-limited API — including ours.
There's a structural reason for that last point which is worth making explicit, because it applies to every vendor in this category. Pinterest's own analytics don't have to ask Pinterest for permission. They read the data directly. Every third-party tool — ours included — is on the outside, requesting numbers 60 times a minute through a public API, caching what it gets, and showing you a copy that is by definition slightly stale.
That's an unavoidable disadvantage, not a bug we could fix with better engineering. So the honest question isn't "whose analytics are more accurate?" — Pinterest's are, always. It's "is there a question Pinterest's dashboard can't answer for me?" Several accounts in one view. Metrics sitting beside the pins you generated. Finding every pin pointing at a dead URL. Those are workflow questions rather than accuracy questions, and they're the only ones worth paying for.
Mistakes that waste analytics
Treating a zero as a fact
It might be a window that predates the pin, or a throttled request. Fix: check the pin's age and whether the fetch actually succeeded before you conclude anything.
Tracking "clicks" as one number
Pin clicks and outbound clicks measure different behaviours, and the gap between them is the most useful diagnostic you have. Fix: track them separately, always.
Chasing an industry benchmark CTR
There's no sourced number, and "CTR" isn't even one metric on Pinterest. Fix: compare your pins to your other pins.
Deleting low performers because you read that you should
No published Pinterest statement says dead pins hurt you. Fix: delete pins with broken destinations; leave the merely unremarkable ones alone and go make new ones.
Judging a pin in week two
Pinterest distributes gradually — a pin can take weeks to find its audience. Fix: give pins months before calling them failures, and set your "old enough" threshold accordingly.
Optimising impressions
Impressions rising while outbound clicks stay flat means you're growing a number, not a business. Fix: make outbound clicks the metric that matters.
Buying a tool to look at 30 pins
Pinterest's dashboard does that for free, better. Fix: buy tooling when volume is genuinely the problem — not before.
Frequently asked questions
What Pinterest metrics does PinLaunchr show?
The four organic metrics Pinterest's API exposes per pin, plus a few it returns alongside them:
- Impressions — times your pin appeared on a screen.
- Saves — times someone saved it to their own board.
- Pin clicks — taps that opened the pin itself (a closeup, not your site).
- Outbound clicks — taps that left Pinterest and landed on your site.
Pinterest also returns profile visits, follows, reactions and comments on some responses.
The distinction that matters most is pin clicks vs outbound clicks. They sound alike and mean completely different things: one is someone looking closer, the other is someone arriving on your website. If you only track one number, track outbound clicks — it's the only one that becomes a customer.
Why do my Pinterest pins show zero impressions?
Usually one of three reasons, and only one of them means the pin failed.
- The pin is newer than the analytics window. If a report starts 90 days ago and the pin is 10 days old, Pinterest returns no data for the period before it existed. The pin looks dead; it's just new. We hit this exact bug ourselves and fixed it by clamping each pin's window to its creation date.
- Rate limiting. Pinterest's analytics endpoints allow 60 requests per minute. A tool asking for 300 pins at once gets throttled, and a badly written one turns those failures into zeros.
- The pin genuinely got no impressions. This is the boring, real one — and you can't identify it until you've ruled out the first two.
The lesson generalises beyond us: a zero in a Pinterest tool is not automatically a fact. It can be an artefact of how the data was fetched.
Can any tool show true lifetime Pinterest analytics?
Partially, and the constraint is Pinterest's — not any vendor's.
The per-pin date-range analytics endpoint is capped at a 90-day range, and its start date can't be more than 90 days back. Ask for a longer window and it errors. That means no tool can build true lifetime metrics out of that endpoint — you can't chunk your way to two years of history because older windows are simply refused.
Lifetime numbers come from a different call — the pin object itself, requested with metrics attached — which returns a lifetime interval. That's the only path to >90 days, and it carries its own limit: lifetime metrics are available for pin formats created on or after 20 March 2023. Older standard pins don't expose it.
So if a tool promises complete lifetime analytics on a pin you published in 2021, ask how. The API we all share doesn't offer it.
What is a good CTR on Pinterest?
We're not going to give you a number, and you should be suspicious of anyone who does.
There's no published Pinterest benchmark we can point to, and the figures that circulate have no methodology behind them — the same folklore problem as the "45-60 day" seasonal rule. A "good" rate for a niche craft blog and a mass-market retailer aren't comparable, and comparing yourself to an invented average tells you nothing.
There's also a definitional trap: "CTR" on Pinterest isn't one number. Impressions → pin clicks and impressions → outbound clicks are different rates measuring different things. Someone quoting a benchmark usually hasn't said which one they mean.
The useful comparison is you against you. Your own pins, same account, same niche, same audience. Which of your pins earns outbound clicks at a better rate than your others? That's answerable, it's real, and it tells you what to make more of — which is the only reason to look at the number.
What's the difference between pin clicks and outbound clicks?
The single most useful distinction in Pinterest analytics, and the one most people never learn.
- Pin click: someone tapped your pin and opened it on Pinterest. They're looking closer. They have not visited your site.
- Outbound click: someone tapped through to your destination URL. They're on your website now.
Every pin click is interest. Only an outbound click is traffic.
The gap between them is diagnostic. Lots of pin clicks, few outbound clicks means your image earned attention but your promise didn't survive the closeup — the title or the offer isn't compelling enough to leave Pinterest for. That's a copy problem, not a design problem, and you'd never spot it if you were tracking "clicks" as one number.
How does PinLaunchr find underperforming pins?
It syncs your pins into a database, fetches each one's metrics, and flags the ones that did nothing. The rules for what counts as "safe to select" are strict and worth stating exactly, because this is a feature that deletes things:
- Zero saves, and
- Zero pin clicks, and
- Zero outbound clicks, and
- Impressions at or below your chosen threshold, and
- Old enough to have had a fair chance
All five, together. A pin with a single save is not auto-selected.
And one more condition that matters more than the rest: if we couldn't fetch real metrics for a pin, it is never marked safe — because zeros that came from a failed request look identical to zeros that came from a failed pin. Those get flagged for manual review instead. See the underperforming section.
Will PinLaunchr delete my pins automatically?
No. Never. Deleting a pin is irreversible on Pinterest, so the flow is deliberately slow:
- Sync — your pins are mirrored into a database with their metrics.
- Review — you look at what's flagged, with the numbers in front of you.
- Queue — you choose which ones to line up.
- Confirm — a second, explicit confirmation before anything is deleted.
There is no setting that turns this into a one-click purge, and we're not going to add one. A tool that auto-deletes based on metrics it fetched over a rate-limited API — where a throttled request looks exactly like a dead pin — is a tool that will eventually delete something that was working.
The safety rule in the code says it plainly: a pin whose metrics we couldn't verify is never auto-marked safe, because the zeros could hide genuine performance.
Should I delete underperforming pins at all?
Honestly: often no, and we'd rather say that than have you use the feature badly.
Pinterest is a search engine, and a pin that did nothing for six months costs you nothing to keep. It isn't crowding anyone out. There's no published Pinterest statement that dead pins harm your account — and we're not going to invent one to justify a feature.
Where cleanup genuinely helps is narrower: pins pointing at pages that no longer exist (a sold-out product, a deleted listing, a 404). Those actively waste the clicks they earn and give someone a bad experience with your brand. That's worth fixing — though often by fixing the link rather than deleting the pin.
The other honest case is a tidy-up after a strategy change, where old pins misrepresent what you now do.
If you're deleting pins because you read that low performers hurt you, stop and go make new pins instead. That's the activity with actual upside.
Why is per-pin Pinterest analytics so slow in every tool?
Because Pinterest's rate limits make it physically slow, and this is worth understanding before you blame any vendor.
The analytics endpoints sit in a category allowing 60 requests per minute. One request per pin means a 300-pin account needs five minutes of continuous, perfectly paced calling — assuming nothing else is happening on your account.
By contrast, fetching the pin object itself sits in a category with far more headroom (1,000 per minute). That's a 16× difference, and it's why a well-built tool prefers the pin-object route for lifetime metrics.
Our own tab paces itself deliberately: one pin at a time, capped just under the limit, with a full-minute pause if Pinterest pushes back. It looks slower than it "should" be. That's the correct behaviour — the alternative is a burst of requests that gets throttled and returns a screen full of zeros that aren't true.
Are Pinterest's own free analytics good enough?
For most people, yes — and this is not a comfortable thing for us to publish.
Pinterest's built-in analytics are free, first-party, always current, and never rate-limited against you. They show impressions, saves, clicks and outbound clicks, and they're the source of truth. If you have one account and want to know what's working, open Pinterest.
A tool earns its place at a narrower job: many accounts in one view, metrics sitting next to the pins you generated, finding dead links at scale, or answering "which of my 400 pins should I stop making more of?" without opening 400 tabs.
That's a real problem at real volume. It is not most people's problem, and if you have 30 pins on one account, the free option is genuinely the better tool. We'd rather you knew that than churn in six weeks.
What should I actually do with these numbers?
Three diagnoses, each pointing at a different fix. This is the whole value of analytics — everything else is a vanity dashboard:
- Impressions up, saves and clicks flat → Pinterest is showing your pin and people aren't interested. The image or the promise is wrong. Change the hook, not the volume.
- Saves up, outbound clicks flat → people like the idea but won't leave for it. Your title is aspirational and your offer isn't. Fix the title.
- Outbound clicks up, nothing happening on your site → the pin over-promised, or the page doesn't deliver what the pin implied. The problem is off Pinterest.
Notice none of these is solved by posting more. Analytics are for changing what you make, not how much. If a number doesn't change a decision, it's decoration.
Does PinLaunchr show Google Analytics data too?
There's a separate Website Analytics card that connects a GA4 property per site, which answers a question Pinterest can't: what happened after the click?
The split is worth understanding, because the two never quite agree:
- Pinterest analytics — impressions, saves, pin clicks, outbound clicks. What happened on Pinterest.
- GA4 — sessions, engagement, conversions. What happened on your site.
Expect the numbers to differ. Pinterest counts an outbound click when someone taps; GA4 counts a session when a page loads. People tap and leave before the page renders, so outbound clicks are almost always higher than sessions. That gap is normal — if it's enormous, your landing page is probably slow, which is a genuinely useful thing to learn from two tools disagreeing.
Why don't you publish benchmark statistics?
Because we couldn't verify them, and the rule applies to us too.
Across this blog we refuse numbers with no published methodology — competitors' marketing claims, Pinterest's own advertising statistics, and folklore like "fresh pins get 90% of your traffic." We even refused a competitor's statistic that supported our own argument, because a standard you drop when the number helps you isn't a standard.
The same applies here. We have no way to publish a trustworthy "average Pinterest CTR" — we'd be aggregating our own customers, with our own selection bias, using a definition of CTR we chose. That number would be marketing dressed as research.
What we do publish are Pinterest's API constraints — the 90-day cap, the 60-per-minute limit, the 2023 lifetime cutoff. Those are checkable facts about the platform, they bind every tool including ours, and they're the reason your analytics behave the way they do.
What are the limits of PinLaunchr's analytics?
Stated plainly, because you'll find them anyway:
- Everything Pinterest's API limits, we're limited by. The 90-day range cap, the 60-per-minute analytics rate, the March 2023 lifetime cutoff — those aren't our choices, and no vendor can engineer around them.
- Per-pin analytics is paced, so it's slow on large accounts. Deliberately.
- We're Pinterest-only. No Instagram, no TikTok, no cross-network dashboard.
- We're not a listed Pinterest Marketing Partner. Tailwind, Later, Buffer and Hootsuite are.
- No permanent free tier — a 7-day trial with 30 credits. Pinterest's own analytics are free forever.
The honest summary: our analytics exist because they sit next to the pins you generated and cover several accounts at once. If that's not your problem, Pinterest's own dashboard is free and excellent.
Wrapping up
Pinterest analytics look simple and aren't, and most of the strangeness you've seen has a cause: a 90-day cap on the range you can request, 60 analytics requests a minute, and lifetime metrics only for pins created since March 2023. Those bind every tool on the market. They're why your reports are slow, why lifetime numbers are patchy, and why zeros show up where they shouldn't.
The four numbers worth your attention are impressions, saves, pin clicks and outbound clicks — and the last two are not the same thing. Only outbound clicks are traffic. The gap between them is the most useful diagnostic Pinterest gives you, and collapsing them into "clicks" throws it away.
Our analytics exist for a narrow reason: they sit next to the pins you generated, across several accounts, so "which of these 400 should I stop making?" is answerable without 400 tabs. If you run one account, Pinterest's own free analytics are better than ours — always current, never rate-limited against you, and free forever, which we're not.
And whatever you use: a zero is not automatically a fact, a benchmark without a methodology is not evidence, and the only reason to look at any of this is to change what you make next. If a number doesn't change a decision, it's decoration. When the answer is "make a lot more, in more angles," generating them in bulk is the part we built.