Emanuel Cinca was spending $1,000 a month on reporting. Getting clean data from his newsletter stack required manual exports, pivot tables, and hours of work just to get a straight answer.
He got to the point where he needed to figure out a different, more efficient way to report.
He spent 30 hours building his own analytics dashboard with AI as his development partner. The total cost today: under $10 a month in hosting fees.
Here's a step-by-step guide to building it yourself:
π€ The problem he was solving:
Clean subscriber data requires reconciling multiple sources, UTM parameters, churn calculations, cost per engaged subscriber, not just initial sign-ups. Without easy access to that data, decisions slow down. As Emanuel puts it: "If you have to put in too much work to get that data, you're less likely to use it frequently. So you make less informed decisions."
π What the dashboard revealed:
Once it was running, the data showed something he hadn't been able to see clearly before. His generic paid campaigns were churning new subscribers at 40 to 45%. His more targeted campaigns were churning at around 30%. That gap had always existed. He just hadn't had fresh enough data to catch it in time to act on it.
π¦Ύ How he built it:
He had minimal coding experience. His approach was to describe the problem to AI in plain language, ask it to confirm whether a solution was possible, and then build step by step with AI handling the code and guiding the infrastructure setup. Google Cloud for hosting, Python for the backend, a web-based frontend his whole team could access through their existing Google Workspace login.
π His biggest lesson:
Be extremely specific. "It's better to repeat yourself and say too much rather than not enough." Vague inputs produce vague outputs. Name the exact field, the exact date parameter, the exact metric every time.
β
The result:
The dashboard now processes thousands of subscriber interactions daily, calculates true cost per lead by accounting for early unsubscribes, and tracks churn by campaign automatically. $1,000 a month became under $10. And the decisions that used to wait for a Friday reporting session now happen in real time.