I've been using n8n for a few months now, and it's replaced a surprising number of paid services I used to rely on. Between expense tracking apps, social media schedulers, news aggregators, and various productivity tools, I was spending around $40 a month on subscriptions that all did one thing — automate simple tasks. So I built my own automations instead.

Finance tracker

No need for a paid finance tracker that goes through my emails and messages

I've always wanted an expense tracker that doesn't need access to my messages and emails. Most apps require permissions to scan your inbox, categorize transactions, and build spending reports. It works fine for online purchases, but it can't track cash, and I'm not comfortable handing over that much access to my data.

I replaced all of that with a simple n8n workflow. Whenever I spend money, I send a quick message to Telegram like "coffee, 3 USD, today" or "groceries, 45, yesterday." An AI parser extracts the amount, date, and category, then logs it automatically into a Google Sheet. I can phrase it however I want, and the AI still understands.

The setup took maybe thirty minutes. A Google Sheet stores the entries, a Telegram trigger starts the workflow, and a sub-workflow handles the parsing and logging. The only tricky part was making sure the AI Agent pointed to the correct sub-workflow, but once that was done, it's been running smoothly without any maintenance. You can use OpenAI's GPT-4, Claude, Gemini, Deepseek, or whatever AI model you prefer.

Personalized tech newsletter

Save costs on subscriptions like Feedly Pro

I follow a lot of tech news, but keeping up with it is impossible. I wanted a way to stay updated without being glued to a feed all day, so I built a weekly tech newsletter that writes itself. This automation collects, filters, and summarizes news from top tech sites and turns it into one clean email that lands in my inbox every Friday morning.

Here's how it works. Every morning, a scheduled trigger fetches articles from RSS feeds of sites I want to follow. The workflow extracts the title, summary, and publication date, then converts everything into vector embeddings using OpenAI. This lets the system search for and prioritize stories semantically instead of just by keywords.

Throughout the week, all these stories are stored temporarily. On Fridays, a separate scheduled flow activates. It queries the stored data for the most relevant stories based on my chosen topics (AI, hardware, games, and new gadgets), summarizes each article, and rewrites them in a natural, easy-to-read tone.

The setup is straightforward. You connect your OpenAI and Gmail accounts to n8n, update the list of RSS feeds you want, and define your topics of interest. Once everything is connected, the automation runs quietly on its schedule. If you prefer Telegram instead of email, you can easily swap the final step.

Social media cross-posting

You don't need Buffer or Later

I've recently started experimenting with short videos, just testing the waters by posting clips across different platforms to see what sticks. But managing six different apps and their clunky upload tools was a nightmare. Every platform has its own quirks, formats, and login steps. I tried Buffer for a while, but the free plan was restrictive and the paid version felt overpriced for what I needed.

So I built my own posting setup with n8n. It does exactly what Buffer or Later does, except it costs nothing and runs entirely on my own server.

The workflow starts with a simple form where I upload content (text, images, or videos) and select which platform I want it to go to. I pick from Instagram and TikTok, add a caption, choose the profile, and upload the file. Once I hit submit, the workflow handles everything automatically using a third-party API service called Upload-Post.

When I submit the form, n8n captures all the inputs and checks whether the upload is a photo or a video. If it's an image, it routes it through the photo upload API. If it's a video, it goes through the video API instead. Both routes send the caption, account name, and platform details to Upload-Post's server, which handles publishing.

Once the upload is done, n8n parses the response and confirms whether the post went through successfully. If something fails, the workflow logs the error automatically so I can fix it later.

YouTube video summarizer

When you can't sit through a one-hour podcast

I watch a lot of long-form videos, mostly for research and learning. The problem is that I don't always have time to sit through an hour of content just to get a few important takeaways. I wanted a quick way to understand what a video covers before deciding whether it's worth watching.

I built a workflow that does the heavy lifting for me. All I need to do is paste a YouTube link into a simple interface, and within a few minutes, I receive a clear summary with the main topics, key insights, and supporting details.

The workflow starts with a Webhook node as the entry point. Once I submit a video URL, a YouTube API node fetches the title, duration, and description. Then, a Transcript Extraction node retrieves the full transcript. It works with most public videos and supports different link formats, including shortened ones.

Once the transcript is ready, it's passed to a GPT-4 node that processes the content. The AI identifies the structure of the discussion, organizes it into sections, and rewrites the transcript into a readable summary.

I have two ways to receive the final summary. The first option sends it directly to my Telegram so I can read it on my phone. The second option logs all summaries into a Google Docs file that serves as an ongoing research notebook.

Recurring subscription audit and cancellation alert

Track all the subscriptions

I kept forgetting which subscriptions were renewing and when. Between streaming services, tools, and random software trials that quietly turned into paid plans, I was wasting more money than I realized. I used to manage this manually in a spreadsheet, but it took hours every month and was easy to forget.

Now, n8n handles the entire process automatically. The workflow runs every morning at eight and checks my subscription data stored in a Google Sheet. It looks for upcoming renewals, due payments, or services that have recently increased in price. The results are formatted into a short summary and sent directly to me on Telegram.

If I reply with something like "cancel Spotify" or "update Netflix to yearly," the workflow understands it. It uses a GPT-4 node to interpret the message and make the appropriate change in the Google Sheet. I can also add new subscriptions or update existing ones through the same chat.

You don't need a subscription for everything

I'm not saying n8n is the perfect solution for everyone. It takes time to set up, and you need to be comfortable with basic automation concepts. But if you're already paying for multiple services that essentially automate simple tasks, it's worth considering whether you could just build those workflows yourself.

The possibilities are unlimited. You can build a self-healing home lab with n8n, automate your inbox with an email cleaner that keeps it clutter-free, and even set up a few micro-automations that take less than five minutes to configure.