Quick reference from 120+ projects. Which tool to use, when to avoid each one, and how fast it pays back.
| Tool | Platform cost | Use when | Avoid when |
|---|---|---|---|
| Apps ScriptFree | $0/mo | Business runs on Google Workspace. Data lives in Sheets, Drive, or Gmail. Per-run cost is a concern. Logic is complex enough that a flow builder gets messy. | Sub-second response needed. High concurrency. Mission-critical uptime. |
| Zapier | $20–$500+/mo | You need 10+ SaaS-to-SaaS connections quickly. Non-technical team will maintain it. Volume is low enough that per-task costs stay manageable. | High volume (costs compound fast). Anything where silent failure is unacceptable. |
| Make.com | $9–$200+/mo | Complex multi-step visual logic. Team comfortable with flow builders. More flexibility than Zapier at lower cost. | Same silent failure risk as Zapier. Same volume cost problem. |
| AppSheet | $5–$10/user/mo | Mobile app for field teams. Form-driven, data-centric workflow. Data already in Google Sheets. No custom UI required. | Client cares about how it looks. Complex interactions. Custom branding requirements. |
| n8n | ~$0 (self-hosted) | Technical team wants full control. High-volume workflows. Cost-sensitive at scale. | Non-technical clients. Fast delivery is the priority. |
| Custom build | Hosting only | Real-time processing required. Zero-downtime SLA. Volume that makes any tool cost painful. Workflow complexity has outgrown visual tools. | Most businesses. The bar for "we need custom" is higher than most people think. |
When a Zap or scenario breaks, it fails quietly. No alert. You find out days later when downstream data is wrong. Build your own error alerting if running anything critical.
Apps Script, Zapier, and Make all struggle with sub-second latency, high concurrency, and zero-downtime requirements. These are fundamental architectural limits, not bugs.
AppSheet's UI customisation has a hard ceiling. If a client has specific layout, branding, or interaction requirements — clarify this before scoping. The data model can be perfect and the client still unhappy.
6-min execution cap, daily quota limits, no persistent server. Nearly all of these have workarounds within Apps Script itself: chained triggers, PropertiesService, LockService, batching.
App formulas run on the device — instant. Computed columns and workflow rules trigger a server sync — laggy. Move calculations to App formulas wherever possible. It's the most common performance fix we make.
"Can I write the rules for this?" If yes — even if the rules are complex — rule-based automation is almost always better. AI belongs only where input is genuinely unstructured and rules break down.
| Current monthly platform cost | Typical build cost | Break-even | Year-1 net saving |
|---|---|---|---|
| ~$50/mo | ~$125 | 2–3 months | ~$475 |
| ~$100/mo | ~$125 | ~6 weeks | ~$1,075 |
| ~$200/mo | ~$125–$250 | ~3 weeks | ~$2,150 |
| $500+/mo | ~$250–$500 | Under 2 weeks | $5,500+ |
Build cost = ~5 hrs at $25/hr. Higher-cost rows assume slightly more complex workflows.
"Can I write the rules for this in plain English?"
Rule-based automation. Pick the cheapest tool that connects what you need. Apps Script if it's Google Workspace. Zapier/Make if it's SaaS-to-SaaS.
The input is genuinely unstructured or the variation is too wide to write rules for. This is where AI (Claude, GPT) earns its place — semantic understanding, not just routing.
Send us a description of the process you want to automate. We'll tell you which tool fits — and whether it's even worth automating at all. Honest answer, no sales pitch.
Get a free assessment →