Hey everyone, new here and honestly feeling a bit out of my depth most days! 😅 I’ve been building analytics pipelines at a tiny SaaS startup (think 5 people tiny), mostly stitching together Python scripts, some Airflow, and BigQuery.
I keep seeing all these posts about complex marketing stacks and felt like I had to share this, maybe to get some validation or to be told I'm completely wrong. For context, I'm the entire "data team" and also handle the basic analytics for our marketing person.
So, my possibly unpopular opinion: If you're generating under 50 new leads a day, you probably don't need a dedicated marketing automation platform (like a Marketo or HubSpot's full suite). The overhead is huge for the volume.
Here's what I cobbled together instead that's been running (mostly) okay for months:
* Lead forms write to our Postgres DB (via the app backend).
* A simple Python service (I used FastAPI) tags and scores them based on source/activity.
* A daily Airflow DAG pushes a lead list to a Google Sheet for the sales team.
* Basic email sequences are handled by… well, Mailchimp's free tier, using their API. The segmentation logic is just a few `if` statements in my Python script.
I spend maybe an hour a week maintaining this. The main pain point is when the API calls fail (attached a screenshot of my latest Airflow failure 😬). I'm sure it's not "best practice," but it works and the cost is nearly zero.
I'm here to learn how to make these janky pipelines more robust and to understand when it's *actually* time to graduate to a real platform. Am I crazy for doing it this way? Has anyone else gone the DIY route and regretted it later? Looking for stories and maybe some Airflow monitoring tips!
null