Here’s Everything My Substack Dashboard Says
I opened up my Substack analytics - open rates, views, growth, all of it. Now I need two minutes of yours to decide what comes next.
Today, I’m opening up my Substack dashboard and showing you exactly what’s working, what isn’t, and where I honestly stand.
I want to be transparent with you folks. You give me your valuable time and you deserve to know this.
Six months ago, AgentBuild was a few hundred people. Today, it’s 1100+, spread across 73 countries - from Bangalore to Boston to Bradwell.
I want to index on proving value over increasing subscriber count.
Hence, I’m going to ask for your help.
First, the Growth Curve
A year ago, AgentBuild had roughly 300 followers. Today it’s 1,158 - nearly a 4x jump, with the steepest acceleration happening in just the last four months. Something clicked around April, right when I doubled down on decision traces, agent registries, and evaluation frameworks instead of chasing whatever AI news cycle was trending that week.
Where you’re all reading from:
India: 37%
United States: 24%
United Kingdom: 7%
Germany: 3%
Canada: 2%
Plus pockets across 68 other countries
That’s a genuinely global room of practitioners, which still blows my mind every time I look at it.
Where I stand today
My last 20 posts averaged an open rate between 22.5% and 30.3%, with a 30-day rolling average of 25.24%.
Let’s put that in context:
General email marketing average: ~21-22% open rate
Substack’s platform-wide headline number: often quoted at 35 - 37%
Real-world compressed range for niche newsletters in 2025-2026: closer to 28-32%
So where does that put AgentBuild?
Comfortably ahead of generic marketing email, but trailing the Substack averages. I call it “solid, not spectacular.” And that’s fine.
My best-performing piece, “Why Solution Architects Are the Real Force Behind Enterprise AI Transformation,” hit a 30.28% open rate - proof that when the topic lands, the number moves. Which is exactly why I want your input on what topics actually land for you.
My most-read piece by a wide margin: “Why Agent-to-Agent Communication Fails - How to Design for Failure” pulled 897 views, nearly double my typical post.
Close behind: my two-part series on building an AI agent registry (586 and 476 views), and “The Evaluation Graph: Why Your AI Pipelines Are Lying to You” (466 views).
There’s a clear pattern here - you want tactical, failure-mode-focused content over broad conceptual overviews.
Take the survey, please
If I’m grading myself against the benchmarks: good, not great. I’m beating the average email newsletter by a healthy margin, but I’m not yet in “50%+ open rate, readers treat this like a must-open” territory that the best niche Substacks report.
There’s a room to close that gap and the fastest way to close it is writing more of what you actually want, instead of guessing.
Hence, I’ve put together a short survey to answer three things:
what topics you want more of,
what format works best for you,
which of the three community features should I launch
👉 Take the survey here
Your answers will directly shape the next quarter of AgentBuild issues. This isn’t a formality - I’m genuinely building the next stretch of this newsletter around what you tell me. I don’t want to make assumptions.
What I’m Building Next
Based on the feedback I’ve already gathered, three things kept coming up as the most wanted next step for this community:
Ask Me Anything sessions - live, unscripted Q&A where you bring your hardest production AI problems and I answer in real time.
Virtual meetups - smaller group calls connecting practitioners across India, the US, UK, and everywhere else you’re reading from.
A reader community forum - a space to swap notes and debate frameworks between issues, not just consume them passively.
I’m committing to build all three.
The only question left is sequencing - which one should I do first?
Tell me in the survey above.
Thank you for being part of the community. I am focusing on delivering value to each of you. Feel free to reach out if you need any help.
Talk soon,
Sandi
👉 Take the survey here
P.S. If you’re new here - welcome 🎉. AgentBuild is a community of practitioners working through the real challenges of getting AI into production inside large organisations. Every week I share practical, grounded thinking from the people doing this work at the sharp end. The goal is never theory - it’s always: what can you use Monday morning.
Ask your friends to join.
More valuable content coming your way.