What Actually Counts as Good? Honest Pinterest Benchmarks by Account Size

Somebody in a group I lurk in asked whether ten thousand monthly impressions was good. Eleven people answered before I got there. Four said yes, three said no, two said it depends, and the last two asked what niche she was in — which was the closest anybody came to being useful.

I typed out a confident, encouraging answer about how ten thousand is a perfectly respectable start. Then I deleted it, because it was the same answer I had given somebody eighteen months earlier who turned out to be doing badly, and the same answer I had given somebody else who turned out to be doing brilliantly. An answer that fits both cases is not an answer.

The honest response is that the question is malformed — not to be difficult, but because the number she gave me genuinely does not contain the thing she wanted to know. Here is why, then the bands I actually use, which are wide on purpose, and then the three ratios I would swap the whole lot for.

Why “is ten thousand impressions good” has no answer

An impression count is really three things multiplied together: how many pins you have published, how long those pins have been out there compounding, and how well any individual one performs.

Change the first and the third can halve while the total still goes up. That has happened to me, in public, with some enthusiasm. I had a month where I put out roughly sixty pins instead of my usual twenty, watched impressions climb about seventy per cent, and quietly concluded I had worked something out. What had actually happened was that each pin was performing considerably worse than the month before and there were three times as many of them.

It took me until the following month — when I went back to twenty pins and the total collapsed — to see it. The lesson stuck harder than most, because I had spent four weeks feeling clever about a number that was recording my labour rather than my judgement.

So when two accounts compare impression counts, they are comparing three variables at once and attributing the whole difference to whichever one they were already thinking about. That is not a benchmark. That is a mirror. The same is doubly true of the profile number, which also sweeps in pins from any account you have claimed — why monthly views swing so violently is worth ten minutes if yours has just done something alarming.

The bands I use anyway, and exactly how much to trust them

People still want numbers, and refusing to give any is its own kind of unhelpful. So here are the ranges I see across the accounts I have actually looked at, sorted by age rather than follower count, because age is the variable that drives them.

• A young account — under six months, publishing consistently — usually sits somewhere between a few hundred and a few thousand impressions a month, with outbound clicks in single or low double digits. This looks like failure and almost never is.
• A growing account — six to eighteen months, a few hundred pins live — tends to land in the tens of thousands of impressions, with outbound clicks in the dozens to low hundreds. This is the first stage where the ratios are readable at all.
• An established account — two years or more, a thousand-plus pins working — can be anywhere from a hundred thousand to several million impressions a month, which tells you how useless the top of that range is as a target.
• Saves, at every stage, run somewhere around half a per cent to two per cent of impressions. That band is narrow enough to be genuinely diagnostic, which is exactly why it is the one I watch.
• Outbound clicks run lower still — a fraction of one per cent of impressions is ordinary, and seeing that written down has stopped a lot of people panicking about a number that was perfectly fine.

Notice the shape of that list. The impression bands span two orders of magnitude and the ratio bands span less than one. That gap is the entire argument of this post, and if you stop reading here you have had the useful part.

Ratios compare. Counts do not.

A ratio survives all the things that make a raw count meaningless. Publish twice as much and your impressions double while your saves per impression sits still — which is precisely what you want, because it means the number is describing your pins rather than your workload.

It also survives the event nobody plans for, which is one pin going unexpectedly large. A single outlier can add a zero to your monthly impressions and tell you nothing repeatable about anything. It moves a ratio far less, and on the rare occasion it does move one, it moves it in a direction that means something.

The three ratios worth tracking, and how to compute them

1. Saves per impression. Saves divided by impressions, over the same window for both. Under about half a per cent and something in the pin is not landing; above two per cent and you should be making considerably more of whatever that was.
2. Outbound clicks per pin click. Of the people who opened your pin, how many actually went through to your site. This is the one that catches a pin writing cheques the page cannot cash, and a widening gap here is the earliest honest warning you get.
3. Impressions per pin published. Total impressions divided by how many pins you put out in that period. It is the only version of the headline number that survives a busy month, and it is the one that genuinely tells you whether the account is getting stronger rather than just busier.

Two rules for all three, and both of them have bitten me. Use the same date range on the top and the bottom of the fraction — date ranges quietly change the answer more than anything else on that dashboard, and a thirty-day numerator over a seven-day denominator produces a figure that looks fantastic and means nothing.

And never compute any of them for a single day. Pinterest data settles over several days, weekends behave differently from weekdays, and a Tuesday on its own is noise wearing a number’s clothes. If the denominators themselves are the confusing part, the full glossary of what each number counts sorts that out in one sitting — particularly the impressions and reach distinction, because quietly using reach where you meant impressions changes a ratio by a factor of several and nothing warns you.

How long before any of this means anything at all

Longer than you want. Pinterest is slow in a way that other platforms simply are not — a pin can start earning six months after you publish it, which is lovely for the pin and ruinous for early benchmarking.

I would not read the ratios seriously before about ninety days of consistent publishing, and I would not judge a single pin before roughly six weeks. How long Pinterest actually takes to work covers the shape of that curve properly, and the shape is the point: it is flat for far longer than feels reasonable and then it is not.

Which means the only honest benchmark for a two-month-old account is: are you still publishing. That is genuinely it. Everything else at that stage is measuring the temperature of a cake that is still in the oven, and I have watched three people give up at week seven on the strength of a number that was never going to say anything yet.

Where published benchmarks come from, and why they flatter

For about a year I confidently repeated a figure I had read in a case-study roundup — something like three per cent being an average engagement rate. I stopped the day I worked out where numbers like that actually come from.

They come, almost entirely, from people who had something worth writing up. Nobody publishes the case study titled “I pinned diligently for eight months and it did very little”, and yet that account exists in enormous numbers and belongs in any honest average. The published figure is not a lie. It is a survey of winners presented as a survey of everyone.

There are two further distortions worth naming, and both have caught me out personally rather than theoretically.

• Season, unstated. A wedding account measured in March and the same account measured in November are two different animals as far as the numbers go, and almost no benchmark tells you which month it was taken in.
• One outlier treated as the baseline. A single pin that went unusually wide can carry an entire year of an account’s reported figures, and it invariably gets reported as though it were the steady state rather than a lucky Tuesday.
• Niche mixing. Food and home decor get saved far more readily than almost anything else; finance and B2B considerably less. An average across all of them describes no account that has ever existed.

The only benchmark that is genuinely yours

Your own account, three months ago.

Same niche, same person making the pins, same posting habits, roughly the same seasonal position. Every variable that ruins a cross-account comparison is held constant, and the only thing that meaningfully changed is what you did. There is no other comparison available to you with that property.

So the review I actually run takes about ten minutes, once a month. Pull the same window this quarter and last. Compute the three ratios for both. Write the pairs down somewhere that is not the dashboard — a notebook, a spreadsheet, the back of an envelope — because Pinterest will not remember for you what it said in May, and by August you will have quietly rewritten the memory in your own favour.

What you are reading is a direction, not a level. Saves per impression drifting up means your pins are getting better at their job even if the headline count did nothing all quarter. Impressions per pin drifting down while total impressions rise means you are running harder to stand still — which is the failure mode nobody ever catches, because the big number is going the right way and the big number is the one on the screen.

A rising total with a falling ratio is the most comfortable way to have a bad quarter.

What I do with all of this, in practice

Three numbers a month, compared to my own three numbers from the previous quarter, and nobody else’s. It is less satisfying than it sounds and considerably more useful.

I have stopped reading other people’s screenshots as anything other than entertainment. Not out of cynicism — I simply cannot see the three variables sitting behind the figure, so there is nothing in it I could act on even if I wanted to.

I have also stopped chasing the impression count itself. The longer argument for which metrics deserve to change a decision makes the case more thoroughly, but the short version is that impressions measure how hard Pinterest is currently pushing you and saves measure whether anybody wanted the thing. Only one of those is a report on your work, and what a save is really telling you is the other half of that argument.

If you have appetite for a fourth number, make it engagement rate — computed the same way every single time. There are three different denominators in circulation and choosing one and never moving matters far more than which one you chose, because the whole value of the figure is that it is comparable to your own last one.

The reason I care about any of this is that a benchmark is only worth anything if it changes what you make next. And that decision is never answered by learning you are at ten thousand impressions rather than eight thousand — it is answered by knowing which topics people in your niche are already searching for and which of your pins they are already saving. That is the whole job PinPinterest does, and it is why I would rather spend twenty minutes on demand than an afternoon hunting for a stranger’s number to measure myself against.

The woman in the group, incidentally, had about ten thousand impressions and a little over two hundred saves. That is a two per cent save rate, which is at the very top of the band and genuinely excellent — and not one person in that thread, me included at first, had asked the question that would have told her so.