In July 2026 I built a queue of fifty Pinterest pins for this blog. Image, title, description and destination link, all fifty written and ready. I published eleven of them and then stopped.
Most articles about Pinterest for bloggers are written by people at the other end of that story, with a screenshot of a traffic graph going up and to the right. This one is written from pin eleven. It is a smaller story, but everything in it is something I actually did, including the part where I ran out of patience with the interface, and the part where I realised the pins were advertising a version of this site I was about to delete.
Why the queue lived inside WordPress
The queue is a draft post on this blog. Not a spreadsheet, not a scheduling tool, not a paid app. A single unpublished WordPress post containing one table with five columns: pin number, image URL, pin title, pin description, and the article the pin points to.
That sounds primitive and it was the right call, for a reason that only becomes obvious once you are pasting your twentieth pin. The images were already in the WordPress media library, so their URLs were already there. The destination links are the permalinks of my own posts, so if a slug changes I can see the mismatch in the same place I edit the post. And because it is a post, I can read it out through the REST API and parse the table rows in a script, which is how I can check that the destinations still resolve.
Writing fifty titles and descriptions in one sitting, rather than one at a time in the Pinterest composer, is also the only way I found to keep them from all sounding identical. Side by side in a table you can see when several of them open the same way, which you never notice when you write them one a day.
The three things that actually cost me time
None of them were the things Pinterest advice warns you about.
The destination field lies to you. In the pin builder, when you create a pin by pointing it at an image URL, the destination link field is pre-filled with that same image URL. If you do not notice, you publish a pin whose link sends the reader to a picture on your server instead of to your article. It looks completely normal in the editor. I caught it early, but I have no confidence I would have caught it at pin thirty, at speed, in the evening. Now I check the destination field as the last action before publishing every single pin, out loud, like a pilot.
The composer stops working after two or three pins. Consistently, after publishing two or three pins in a row in the same browser tab, the pin builder would stop responding: the page went blank on reload and nothing in the interface reacted. Reloading did not fix it. Opening a new tab did, every time, and the old tab stayed dead. I never found out why. What that meant in practice is that the job is not “publish fifty pins”, it is “publish three pins, open a new tab, repeat sixteen times”, which is a different and much more tedious job than the one I had planned.
Three a day is a schedule, and schedules need a person. The plan was three pins a day, which is modest by Pinterest standards and still means showing up daily. Pins one to eleven went out over several sittings in early July, the last three on 11 July 2026. Then a week passed, then two, and by the time I came back to it the whole project had been overtaken by the state of the blog itself.
The numbers I do not have
This is the section that would normally hold a chart. I am not going to fabricate one.
I have not pulled a Pinterest Analytics export for this account, and I have not sat down with the referral breakdown in Google Analytics for the period the pins were live. Until I do, I have no impressions, no saves, no outbound click count and no referral sessions to show you, and quoting a number I have not opened would be exactly the habit that got the rest of this site into trouble.
What I can give you is the search side, because I did open that: Search Console’s figure for the three months to 19 September 2026 was fifteen clicks, at an average position of 67. That is the context eleven pins were launched into: a site that was effectively invisible, with no established audience anywhere.
Even with a full analytics export, eleven pins from an account with no following would not prove anything about Pinterest as a channel. The sample is too small and the variables are all moving at once. Anyone presenting eleven pins as a test is selling you something. So the honest summary of my Pinterest results so far is: unknown, and unknowable from this data.
The board name was the real problem
The eleven pins went to a board called “Make Money With AI”. The titles and descriptions matched it: side hustles, income streams, consulting, automation services.
That was faithful to the blog as it existed in July, and that is the problem. The blog as it existed in July repeated the same monthly income promise 108 times across a hundred posts, and had been turned down by AdSense for low value content. The pins were doing their job perfectly: they were advertising, accurately, a promise I could not back up. Building a distribution channel on top of content you are about to disown is work you will throw away, and I threw away this work.
I wrote up what the export of those hundred posts showed in a list of the mistakes behind four rejections, and the money side, including why there are no income figures anywhere on this site any more, in what a small blog can and cannot earn. The order matters: fix what you are pointing people at before you start pointing people at it.
What I would do differently, and what I would keep
- Keep the queue in a draft post. It survived the whole mess intact and it is the only part of this I would not change.
- Name boards after what the reader gets, not after an outcome I cannot promise. “Building small sites with AI” describes a shelf. “Make Money With AI” describes a claim.
- Write ten pins, not fifty, and publish all ten before writing more. Forty finished pins sitting in a draft are not an asset, they are a decision I postponed.
- Check the destination link on every pin, without exception, as the last thing before publishing.
- Open Pinterest Analytics on day one and note the starting point, so that there is something to compare against later. I did not, which is why this post has no numbers.
- Point pins only at posts that are finished and that I would be happy for a stranger to land on cold. Several of my fifty pointed at posts that are now redirected or unpublished.
What happens to the queue now
It gets rebuilt from scratch, because most of it points at URLs that will not exist. Seventy-five of the hundred posts are coming down, with redirects for most of them, so a large share of the fifty destinations would land on a 301 at best. Rewriting the destinations is the easy part; deciding which of the surviving posts is worth a stranger’s first impression is the part that takes the afternoon.
The images have to be redone too. The original pin images were made for that board and that pitch, so they go with the rest of it.
My plan, when the rewrite is finished, is ten pins pointing at ten finished posts, published over two weeks, with the analytics baseline written down before the first one goes out. If there is anything to report after that, it will appear here with the dates and the export it came from. Until then, the only defensible thing I can say about Pinterest is that building a queue is easy, publishing it is tedious, and neither of those is the hard part. The hard part is having something at the other end of the link that deserves the click, which is the whole reason this blog is being rebuilt, a process I am documenting in mapping a hundred posts into clusters and in my launch routine in Search Console.