3
4 Comments

How does Substack ranking algorithm work?

I've been writing a newsletter covering MIT startups. I've grown my subscriber list pretty quickly with little / no marketing, but noticed I am receiving very little traction organically through Susbtack.

https://seamuscassidy.substack.com/

When I search for "startup" in the Substack search bar, I have to scroll forever before my Recap newsletter appears. Same is true when I search "recap." I have to type the full name of my publication ("Startup Recap") to rank first in search.

I assume their relevance algorithm uses some sort of keyword search combined with engagement (likes / comments) and subscriber count. But is there any way I can improve this myself by including key words in my article titles or tagging specific posts with search terms (you can do this on medium)?

on June 15, 2022
  1. 1

    Last time I check it was purely sub count.

    However it seems like it has changed. Your newsletter appears to be on the 7th page for "startup" keyword;

    https://substack.com/api/v1/publication/search?query=Startup&page=7&lastSearch=1655328346468&skipExplanation=false

    If you change the skipExplanation variable to true, API will return the explanations for the reasons why it's listed for that specific keyword. Your newsletter has 7 reasons while the top ones has 10. It's possible to compare and improve. But I doubt it's a big factor. Looking at the other results, the one above has 5 reasons and the one below has 3.

    Further more, there are two keys in the results;

    {
    "rankingDetail": "Launched 3 months ago",
    "rankingDetailFreeIncluded": "Hundreds of readers"
    }
    

    It's hard to tell but by looking at it, it's easy to assume they may care about the growth rate. I'm sorry that I'm busy to dig deeper but it'll be interesting to see what's up. But frankly there is no way to tell without an insider. Having a paid plan, being consistent, using a positive language, word count. All may matter.

    1. 2

      Thanks a lot @Rusted , this is pretty interesting!

      I was curious how the algorithm favors popular publications vs new publications with less followers. The keys above would suggest ranking favors newsletters that have published longer and have more readers. Will be interesting to see how my ranking changes once I hit 'thousands of readers' ( I remember in the old SRP it listed whether a pub had 100s or 1000s, but doesn't anymore). Clearly they segment on this.

      Curious how you queried this, but no worries explaining if you don't have time. Really appreciate the help!

      1. 1

        Hey Seamus, sure thing.

        When you make a search on substack, it's possible to see that query on the web browser's network tab on the developer console. It's a great way to explore undocumented API calls.

        https://developer.chrome.com/docs/devtools/network/
        ^ they probably explain better than me :)

Trending on Indie Hackers
Why Indie Founders Fail: The Uncomfortable Truths Beyond "Build in Public" User Avatar 116 comments I built a tool that turns CSV exports into shareable dashboards User Avatar 94 comments $0 to $10K MRR in 12 Months: 3 Things That Actually Moved the Needle for My Design Agency User Avatar 73 comments The “Open → Do → Close” rule changed how I build tools User Avatar 65 comments I got tired of "opaque" flight pricing →built anonymous group demand →1,000+ users User Avatar 44 comments A tweet about my AI dev tool hit 250K views. I didn't even have a product yet. User Avatar 42 comments