9
15 Comments

How to automate 25m+ Pinterest impressions a month and get $0MRR

Over the past five years, using a series of automation processes, I've grown my Pinterest account to 66k followers and 25 million impressions a month and make a whopping $0MRR. The MRR might be surprising, but I'm currently not utilising the traffic in any way. For the most part, this has been a journey of automation and experimentation.

tl;dr
Grew a Pinterest account using Python, Django, Hugo, Git, Airtable and Integromat. It drives traffic to my website. Much of it is automated. I spent a lot of time and effort to end up with a relatively simple solution.

I love automation; I try to automate everything. I'm happy to spend a day creating a system to automate a process that would have taken an hour to do manually. If there is a way to get a positive benefit with no or limited continual input, I want to know about it. I've applied this thought process (perhaps to my detriment) to grow my Pinterest account.

Early days - Instagram straight to Pinterest

This was the fastest way to get content onto my Pinterest account; it was built right into IG. I wish I had thought this through a bit better because all I achieved was getting some new Pinterest followers and sending traffic to Instagram. Acceptable for Pinterest growth, but sending traffic back to IG doesn't give you much control over your viewers. Nevertheless, this is how things rolled for about 12 months.

Getting smarter, but not really

Seeing that I was getting a bit of growth, I wanted to try and control a bit of the traffic my pins were generating. I built a WordPress site, put up a couple of blog articles, added a couple of print on demand products and slid them in between some of the automated pins coming from IG. It had minimal impact on website traffic, and I made maybe one sale a month. Content going up on my site wasn't compelling enough and wasn't frequent enough. It would have worked if I was willing to write more, but I wasn't.

Creating the unnecessarily complicated pipeline

To capture the traffic sufficiently, I needed to own the destination all my pins were linking to. Around the same time, I was deep into learning Python and Django. I convinced myself I needed a Django website. I definitely did not. After a lot of effort, I had a Python script that scraped my IG posts then posted them to my site. Once the posts were live on my site, another script connected to the Pinterest API and posted them to my account. I was very proud of what I built, not so proud of the monthly AWS bill.

The site didn't offer much value other than being able to filter my curated collection of images. Honestly, it's like a low rent Pinterest specifically for imagery I like. But it was something I could control.

The site was complex, slow, and making layout changes was a real pain. I could have achieved the same result using WordPress + Python in far less time for much cheaper. It worked for roughly two years without any significant problems. The site also started generating ok traffic.

No more IG, and a dumb idea

By this time, around three years into my journey, my Pinterest account had a catalogue of 6000+ created pins and a follower count of nearly 25k. I was growing more on Pinterest than on IG, with less effort. IG was also restricting my reach, so I got frustrated, and I stopped posting to Instagram.

The website still needed fresh images, and the Pinterest browser extension was pretty handy at collecting content. However, if I used the extension to post directly to Pinterest, I cut the website out of the equation. I created a system that went: feeder Pinterest account > website > main Pinterest account. I made this work using the browser extension and some Python scripting. Looking back, it seems like an idiotic idea. However, I'm sure, at the time, it made sense and probably required the least amount of effort.

More speed, less cost and getting hacked

All my pins were directing traffic to my site, but the website was slow and bounce rates were high. I was also spending too much on AWS hosting (~$35/month). My solution was to go static and host it for $0. Achieving this was no small feat because I now had a site with 5000+ pages. The only static site generator I could find that would handle this many pages without an insanely large build time was Hugo. I knew nothing about Hugo or the Go language, so of course, this was what I went with. After a few false starts, I found Hugo to be a fantastic framework. It is now my go-to for any content-based website.

Hitting about 90% complete with the Hugo site rebuild, my Django site was hacked. At least, that's what it looked like on the front end. Someone replaced all the images on the front page with a request for Bitcoin donations. It looked terrible. What happened was I had insufficient restrictions on my AWS bucket, and someone was able to replace the images—quite a boneheaded error on my part. I was able to push the Hugo site out pretty quickly but had no automation processes in place to make updates. However, the website was quick, and I was hosting it for free on Netlify.

This is how it sat for 12 months. I made no updates to either my site or my Pinterest account. However, due to the extensive library of pins I had, the site kept a constant stream of traffic (~12k uniques a month and ~25k page views), and the Pinterest account was getting ~3k new followers a month.

Getting it fired back up

My Pinterest > website > Pinterest pipeline only needed minimal tweaking to fit my Hugo site. Pinterest was now also allowing automatic posting by RSS. I refactored my existing Python script and changed it to create Hugo content files. GitHub actions helped me here. I was able to keep all contained one platform and run my script for free. I felt very clever. My pipeline was working, the site was fast, new content was going up, and it cost me $0 a month.

A no-codeish solution

At some point, Pinterest made changes that broke my script. Time to revise the whole system. Enter Integromat and Airtable.

Using Airtable web clipper, I now curate content and store it in a table. Integromat grabs this content, processes the info, creates new Hugo content files and pushes them to GitHub. Pinterest's auto-publish grabs the content from my website and pins it to the appropriate boards. It's a much simpler solution and works way better than anything I had set up previously.

I say no-codeish because Hugo isn't a no-code framework. There is a learning curve, and it can be pretty intimidating to operate if you have no dev experience. However, the actual process to get new content up on my site is all done with Airtable and Integromat. Pinterest takes care of the rest.

Lessons learned

I pass off procrastination as education. The Django website was wildly unnecessary. It was an overly complex solution that I didn't need but felt would be a good learning experience. I have not used Django since and cannot see myself using it again soon. A WordPress site would have taken me 1/10th the time to set up and probably worked better.

I like having a project; I'm scared of having a product. It might be a fear of failure, but I enjoy working on something more than running it. The easy solution was never my first choice. If you continually add features and complexity, you can extend your launch date indefinitely. Being aware of this has helped me infinitely with all new projects I take on.

You need an end goal in mind if you want to make money. I have grown something I'm very proud of, but it makes $0 MRR, and the site has low content value. The main reason for this is I had no goals other than connecting a site to Pinterest. A proper plan for monetisation would have changed the way I approached the whole thing. I would have been more efficient with my time and put less focus on non-essentials.

You can't control anything you don't own. Any social platform can make changes at any time that torpedo your engagement and reach. Relying on a service or platform as your single source of engagement or revenue is a gamble. IG cut my reach by roughly 70% over 12 months. Pinterest could very well do the same. Hopefully, before that, I'll have a proper plan for my website.

Pinterest is a slow burn but well worth the effort. Patience seems to be the key. Keep posting regularly, and your following will grow. Pins also have an extremely long shelf life. I get clicks to my website from pins that are well over 12 months old.

Next steps

I see the potential in starting a blog and selling digital resources. Growing an email list would be part of this process too. However, until I have the time and headspace to dedicate to formulating a proper strategy, I will be satisfied with getting a little serotonin boost from looking at my Pinterest impressions.

Update below

Links

I debated whether I should put links to the site and Pinterest accounts because I've operated them anonymously for so long. I've also received purchase offers on the IG and Pinterest accounts. So if I sold, I didn't like the idea of being linked publicly if they started to promote sites I didn't like. Then there is the phishing, so much phishing. I was getting emails or DMs daily from people trying to access my IG account for a while. But it's probably time to put it all out there.

typegang.com
Pinterest
Instagram this is rarely used now

  1. 1

    Yeah, I'm always a fan of going after the channels that aren't at the top. Top channels = good results. While underutilized ones = CRAZZZYYY results. Good going man

    1. 1

      Thanks!

      Yeah, it seems to be an unusually small group that really pushes Pinterest.

      1. 1

        Nicee. Hey I've been working on a super in-depth guide on how startups can gain a competitive edge. When I post it on here, it'd be great if you could give it a read and share your thoughts. Follow me so you'll know when it drops.

  2. 1

    Hi KFR3000, since you love making things more efficient have you considered making automations for people as your product :) I am just starting a business and using Pinterest and getting some advice from you would be super helpful. If you ever want to share some wisdom or see if I can be your target group let me know :)

    1. 1

      I'm more than happy to help. Maybe one-day ill look at some form of automation as a service model. Right now, I need to force myself to focus on one thing. I suffer from shiny object syndrome. If you have twitter send me a DM and let me know how I can help.

      1. 1

        Hey again. Wow that's great. I dont use Twitter any more, is there another way I can contact you? Maybe via pintrest messages :)

        1. 1

          I sent you an email. Let me know if it doesn't come through.

  3. 1

    What are your IG and Pinterest site?

    1. 1

      Instagram is @type.gang and Pinterest is /typegang

      I've put the links at the end of the article now.

  4. 1

    Great post.

    The only thing I'm missing here is the niche.
    What are the project and images about?

    1. 1

      Thanks!

      It's all in the design niche, I've made an update to the end of the post with links to all the accounts.

    1. 1

      typegang.com

      I've put all the relevant links at the end of the article now.

  5. 1

    Really enjoyed reading this! Yeah definitely, road to growth is made of experimentations, mistakes and sometimes, surprising discoveries. Clever one to focus on Pinterest!

    1. 1

      I'm glad you enjoyed it! Was definitely a journey of discovery, and not particularly linear in terms of output.

Trending on Indie Hackers
I write Lenny's Newsletter, the #1 paid business newsletter on Substack, generating over $500k ARR. AMA. 61 comments Roast my landing page. We help SaaS businesses grow by... 7 comments Considering Cold Mailing? Here's My Personal Experience! 📧 7 comments Do you have 0 paying users? Let's chat 7 comments Tell me About Your Startup! + would love to learn how you use issue trackers + documentation software. 5 comments How can we improve our pre-launch landing page? 3 comments