Archive
Everything published, newest first.
2026
- Paying the API directly versus paying for the app
- The prompts worth keeping in a file
- When a spreadsheet beats an automation
- Twelve automations worth building before any of the clever ones
- AI detectors do not work, and using them is a decision with consequences
- Free tiers that are actually usable, and the ones that are demos
- AI agent steps in automation tools: what they are good for
- An AI built my portfolio page and got three things wrong
- Self-hosting n8n: two weeks in, and the bill in hours
- Deterministic or freeform: two ways AI tools generate output
- We audited our AI subscriptions and cancelled most of them
- Credit-based pricing, and how to work out what it will cost you
- A one-page site in ten minutes: what the workflow really looks like
- Error handling for people who build automations in a browser
- How we test a tool before writing about it
- CV-to-website tools compared on one résumé
- Zapier vs Make vs n8n after a fortnight on each
- AI website generators, tested on the same brief
- The time cost of learning a tool, measured
- The AI tool graveyard: what we adopted and then buried
- AI logo generators are mostly a tax on not knowing a designer
- Notion as the whole workspace, a year later
- When no-code hits its ceiling, and what it costs to notice late
- Use AI to edit, not to draft
- AI image generators for people who are not designers
- Choosing the database under your no-code stack
- Building an internal tool without a developer
- AI slide deck generators tested on a real presentation
- AI meeting notes tools tested on meetings that mattered
- Form tools compared: Tally, Typeform, Fillout and a plain HTML form
- Webhooks explained for people who do not write code
- AI transcription tools compared on difficult audio
- No-code app builders tested on the same internal app
- AI writing tools tested on work we had to ship
- Airtable vs Notion databases: the differences that bite
- Automation that survives a month
- What AI tools actually cost, in time as well as money
- AI writing in real work, not in demos
- No-code stacks that hold up after the demo
- AI generators that produce something you can actually use