A new front desk hire gets a day by day plan instead of a shadowing week and a hope. Day one covers the phone and the diary, day two the intake forms, day three the awkward situations, and by Friday there is a short check that shows what has actually landed and what needs another pass.
Start from Quick Apps Onboarding. It is a live Taskade Genesis app with stages, owners, and progress on every record, which is most of a staff runbook already, though it onboards members rather than coordinators. Move somebody through it, then clone it into your own workspace in about ten seconds and rewrite the stages for your own desk.
The build gives you a plan broken into days with a small number of items each, a link from every item to the app or script that does the job, a buddy assigned per topic, a competency check at the end of each day, a record per hire so the second one is faster than the first, and an AI agent that drafts the content from your existing scripts and processes rather than from a generic template.
The runbook is workable in several shapes:
- Board view with a lane per day and cards moving to done
- List view as the plan the new hire reads top to bottom
- Table view for progress across several hires at once
- Calendar view to see the week laid out with the buddy sessions on it
- Plus the rest of the 7 project views
Automations run the week. On paid plans, Taskade automations release each day when the previous one is complete, notify the buddy before their session, and alert the manager when an item stalls. Across 100+ bidirectional integrations, Slack pushes the daily prompt out while your calendar pulls the buddy sessions in.
The AI agents carry 34 built-in tools including persistent memory. The agent keeps the runbook current: change a script and the onboarding item that teaches it updates with it, which is why the second hire gets a better week than the first.
Workspace DNA is what makes the second hire faster than the first. Memory holds the plan, the scripts it teaches, and how previous hires progressed. Intelligence, drawing on 15+ frontier models from OpenAI, Anthropic, Google, and open-weight providers, drafts the content from your real processes rather than a generic template. Execution releases each day, notifies the buddy, and flags an item that has stalled before it becomes a confidence problem.
Running it in your practice
- Write the week as outcomes, not topics. By Wednesday they can send the right packet unaided. That framing produces a better plan than a list of subjects.
- Clone the app and link every item to a live app. Training that points at the real tool stays current automatically when the tool changes.
- Assign a buddy per topic, not per week. The best person for phones is rarely the best person for forms, and short specific sessions land better.
- Add the end-of-day check. It is short, and it moves the discovery of a gap from week four to day two.
- Update the runbook from the first hire. Whatever confused them is the improvement for the next person, and capturing it takes five minutes.
The practice gets a repeatable week instead of a shadowing arrangement that depends entirely on who happens to be free.
Clone it, invite the team, and the plan is yours. Add a day, drop a topic, or write a short version for a returning staff member.
Where to go next: the front desk call log for the scripts the runbook teaches, the Community Gallery for onboarding boards others publish, or templates for a starting layout. This also pairs with the new patient journey map, which is the best single page to hand somebody on day one.
Frequently Asked Questions
How long should the runbook be?
A week of small items beats a fifty page manual. Three to five items per day, each linked to the real app or script, is the shape that people finish and remember.
Where does the content come from?
From your existing scripts, forms, and processes. The agent drafts from them, so the runbook teaches what your practice actually does rather than a generic version of a front desk role.
What is a competency check?
A short end-of-day confirmation that the person can do the thing: log a call correctly, send the right packet, find an outstanding form. It surfaces the gap on day two instead of week four.
Can we use it for temporary cover?
Yes. A short version covering the phone, the diary, and where to escalate gets somebody useful in half a day, which is exactly what cover needs.
Does it track more than one new hire?
Yes. The table view shows progress across everyone currently onboarding, which matters most when a practice grows two roles at once.
Does the runbook stay current?
It does if it links to the live apps rather than describing them. When a script changes, the runbook item points at the new one, so nobody is trained on last quarter's process.
Does it cover clinical training?
No. This is the front desk role: phones, diary, forms, and coordination. Anything clinical is trained by your practice through whatever route it already uses.
