I thought more automation would save time. It actually created more work to maintain.
When I started automating my digital product business,I had a simple goal:
Remove as much manual work as possible.
The workflow looked straightforward:
Customer buys a product
→ Payment is confirmed
→ Product is delivered
→ Telegram notification is sent
→ Customer is added to Notion
So I connected:
Payhip + Make + Telegram + Notion
At first，it felt great.
A few manual tasks disappeared.
But then I started noticing something I didn't expect.
Every new automation also created something else to maintain.
A webhook can fail.
A field can change.
A scenario can stop running.
A notification can be sent twice.
A customer record can end up duplicated.
And suddenly,I was spending time checking the automation instead of doing the work it was supposed to eliminate.
That changed how I think about automation.
I don't think the goal should be:
"Automate everything."
It's probably closer to:
"Automate the repetitive parts that are expensive to do manually,without creating another system that needs constant attention."
That's what I'm experimenting with now.
I've been documenting some of these workflows as open-source examples using Payhip,Make,Telegram and Notion.
I'm still trying to find the right balance between:
Less manual work vs. less system maintenance.
I'm curious about other indie businesses:
What's one automation you built that genuinely saved you time?
And what's one that ended up creating more maintenance than it was worth?
This is the distinction between automating a task and operating a production system. A useful pre-automation check is: frequency × manual time × error cost should clearly exceed monitoring plus exception-handling cost.
For this flow, I’d add an idempotency key tied to the order/payment, a retry or dead-letter path, and one daily reconciliation view in Notion. Then the system can stay quiet when healthy and surface only exceptions. Which part is creating more maintenance right now: field changes between tools, or duplicate/missed events?
That's a great way to put it.I think that's exactly where the hidden cost comes from — automation forces you to define all the details that a human might normally handle without thinking too much about them.And the more exceptions and edge cases a workflow has, the more maintenance it creates.I'm starting to think the best automation is probably not the one that automates the most, but the one where the rules are simple enough to stay reliable over time.