Good candidates and bad candidates
| Task | Automate? | Why |
|---|---|---|
| Summarising incoming emails and documents | Yes, with review | Low risk, easy to check |
| Drafting client status updates | Draft only | A person edits and sends |
| Creating tasks when a matter opens | Yes | Predictable, rule-based |
| Reminders for upcoming deadlines | Yes, as a supplement | Not a replacement for calendar checks |
| Calculating limitation periods | No | Needs legal judgement and rules |
| Moving client money | No | Regulated; use proper accounting tools |
| Filing with courts | No | Use official systems and human checks |
A staged rollout
- Structure the data: matters, deadlines, actions and time in clear databases.
- Pick one workflow, for example 'new matter creates the standard task list'.
- Run it with a person approving each step for a month.
- Log time saved and errors found.
- Only then loosen supervision, if at all, and add the next workflow.
What Notion offers
Notion has database automations and, since February 2026, Custom Agents that can act on workspace content; see https://www.notion.com/help/custom-agents for what they do, how they are configured and which plans include them. Whether they suit your firm depends on your confidentiality review (see /notion/security-and-confidentiality). For lawyer-oriented examples read /post/custom-ai-agents-for-lawyers.
LawyerOS Pro with AI ($39, one-time) gives the Matters, Actions and Time databases such automations would work on, plus a Custom AI Agents page. It is a template; it does not contain automation software or AI models of its own.
Guardrails checklist
- Written list of approved tools and data rules.
- A named owner for each automation.
- Human approval before anything external is sent.
- A log of outputs and corrections.
- A periodic review of whether it still works after tool updates.
- Check your regulator's guidance on technology and supervision.
Frequently asked questions
- What can a small law firm automate first?
- Creating standard tasks on matter opening and summarising incoming material, with human review.
- Is it safe to automate client emails?
- Draft them automatically if you like, but have a person review and send.
- Do I need coding skills?
- Not for basic Notion automations. More complex integrations may need help.
- Where does automation go wrong?
- On messy data, unreviewed outputs and tasks that need legal judgement.
Sources
About the author
Raghav Handa trained as a lawyer, then spent years building legal technology, from enterprise legal products to custom systems for law firms.
Keep reading
- AINotion Custom Agents for law firms: what they can and can't doWhat Notion Custom Agents are, how law firms can use them for intake triage, deadline digests and matter updates, the costs, and the safeguards lawyers need.4 min read · 30 October 2026
- InsightsThe real cost of DIY AI inside a law firmBuilding your own AI workflows in a law firm looks cheap, but the real costs are time, supervision, security review and maintenance. A framework to count them.3 min read · 17 October 2026
- InsightsWave 2 legal AI without an innovation teamAfter the first wave of legal AI experiments, small firms can adopt practical AI without an innovation team: pick workflows, set rules and review monthly.3 min read · 17 October 2026
- Practice managementA lawyer's productivity guide: protect time, finish the workA practical lawyer productivity guide: capture, weekly planning, deep-work blocks, email batching and reviews, with a simple system you can run in Notion.3 min read · 17 October 2026