Not every automation project is a multi-month platform build with a dozen integrations and a project team behind it. A lot of real, useful automation is much smaller than that: one workflow, one trigger, one clear job to do — save this file where it belongs, send this email when that event happens, log this data without anyone touching a keyboard.
Automations Limited's founder, Mustafa Haider, has delivered dozens of smaller automation builds like these for individual business owners and small teams, mostly using n8n and Make.com, alongside the larger projects featured elsewhere on this site. This page collects twelve of them, grouped by what they actually do. None of them took a large team or a long timeline — each one solved a single, specific bottleneck for the person who needed it solved.
Email & Communication Automation
Email is where most manual busywork hides, and it's also where a small workflow can save the most time relative to the effort it takes to build.
Gmail attachments to a Google Drive subfolder. A recurring need for anyone who receives regular file attachments by email: instead of opening each message and manually downloading and filing the attachment, an n8n workflow watches the inbox and automatically saves incoming attachments into a designated Google Drive subfolder. No more manual downloading, no more forgetting where a file was saved.
Webhook-to-Gmail automated email sending. Built for a case where an external system needed to trigger an email automatically — the workflow listens for a webhook call and sends a Gmail message in response. Getting this working required setting up OAuth2 authentication, a Google Cloud project, and the Gmail API correctly so the emails send securely and in real time, rather than just wiring up a simple send-email node.
Keyword-based auto-reply bot. Uses a Gmail trigger to watch for new emails, then Switch and IF nodes to check the incoming message against a list of keywords — "hello," "price," "support," and similar — and sends back the matching auto-reply. It's a simple structure, but it means routine first-response emails go out immediately instead of sitting in an inbox until someone gets to them.
Gmail to Google Sheets logging. Every incoming email gets logged as a new row in a Google Sheet — sender, subject, body, and date extracted and appended automatically, without ever overwriting what's already there. For a small business without a full CRM, this turns a Gmail inbox and a spreadsheet into a lightweight, searchable record of every conversation.
Real-time email-to-Telegram alerts. Monitors a Gmail inbox for specific keywords — "urgent," "project," and similar flags — and the moment one shows up, sends an instant alert to Telegram through a secure bot connection. Built so the recipient doesn't have to keep an inbox open all day to catch the messages that actually need immediate attention.
Content & Newsletter Automation
Two different projects, same underlying problem: getting a newsletter out the door on a schedule without someone manually assembling and sending it every time.
Newsletter preparation and sending. Automated the recurring steps of putting a newsletter together and sending it to a subscriber list on schedule, removing the manual assembly-and-send routine that had been happening by hand.
Scheduled newsletter from a database. A more data-driven version of the same idea: the workflow pulls records from a database on a set schedule, assembles them into a newsletter, and sends it out automatically — useful for anyone whose newsletter content comes from structured records (in this case, generic transcript data) rather than content written fresh each time.
Data Extraction & Records Management
These three projects share a theme: pulling structured information out of unstructured or scattered sources so a person doesn't have to do it by hand.
Scraping workflows across four websites. Built a set of scraping workflows targeting four separate websites, each pulling structured data automatically instead of someone manually copying and pasting information from page to page. Because each site has its own layout, this meant building and testing four distinct extraction workflows rather than one generic one.
Extracting webpage content from a list of links. Given a list of URLs — in this case, links to meeting and committee agendas — the workflow visits each one and automatically extracts and structures the page content, turning a manual "open each link and read it" task into something that runs on its own.
School management system on Make.com and Airtable. The most involved project in this group. Google Forms collects student, teacher, and class data, which is sent via a webhook trigger into Make.com. From there, Make.com connects to Airtable using a mix of built-in modules and custom HTTP API calls, checks whether a record already exists, and either updates it or creates a new one — upsert logic, rather than blindly adding duplicate rows. The workflow also manages linked relationships between Students, Classes, and Teachers inside Airtable, so the data stays connected instead of living in disconnected tables.
Technical & Research Engagements
Not every engagement was a finished workflow. Two projects were more technical or research-focused in nature.
API documentation analysis for trigger and action definitions. Worked through a set of API documentation to identify and define the available triggers and actions it exposed — the groundwork needed before anyone can build a working integration or workflow on top of that API.
Python development for an AI productivity research study. Contributed as a Python developer — working with Scikit-Learn, Django, and general machine learning tooling — to an enterprise-scale research study on AI productivity. A different kind of engagement from the workflow builds above, but it draws on the same underlying technical foundation.
What This Range Shows
Taken together, these twelve projects span three different levels of automation work: no-code workflow platforms (n8n and Make.com), direct API-level integration (OAuth2 setup, custom HTTP calls, upsert logic against Airtable), and general-purpose development (Python, machine learning tooling). That range matters because most real automation needs don't fit neatly into one category — a single project might need a workflow platform for the easy parts and a direct API call for the one thing the platform can't do out of the box.
It also shows that automation work doesn't have to start big. A single Gmail-to-Sheets logger or a keyword-triggered auto-reply can be built, tested, and running within days, and it can still save real time every single week it runs.
If you have one specific process that's eating up time — something that looks like any of the examples above, or something more custom — it's worth a conversation regardless of how small it seems. Book a free automation audit and we'll tell you honestly whether it's worth automating, what it would take, and what it would cost.