Skip to content

Who does what

Four roles touch the setup. One person can hold all four; in most companies they are three people with ticket queues.

Role What they need Steps Time
Applane admin (whoever signs up) A Google work account at your domain Sign up, create the company, paste the values, run the checker 15 min
DNS owner Access to your domain’s DNS Add one TXT record 5 min, plus the ticket
GCP admin Owner or editor on one GCP project in your organisation APIs, IAM, budget, consent screen, OAuth client 20 min
Google Workspace admin Super admin in the Google Admin console Chrome extension policy, Apps Script check, two identities, Drive sharing 15 min
Builder A Workspace account in the builders group Install, sign in, turn on the Apps Script API once, build 15 min for the first app
# Step Who Time Where Page
1 Sign up and create the company Applane admin 2 min Applane console Sign up
2 Add the applane-verify TXT record DNS owner 5 min + queue Your DNS provider Verify your domain
3 Enable five APIs, grant one role, add a budget alert GCP admin 10 min Terminal or Cloud Shell Set up the project
4 Consent screen (Internal) and OAuth client (Web application) GCP admin 5 min Google Cloud console OAuth consent screen, OAuth client
5 Allow or force-install the extension; create the bot account and admin group Workspace admin 15 min Google Admin console Chrome extension policy, Accounts and groups
6 Paste the project id, client id, group and bot into the console; run the checker Applane admin 5 min Applane console Setup checklist
7 Install, sign in, turn on the Apps Script API Builder 3 min Chrome Install and sign in
8 Build and deploy the first app Builder 10 min The extension Your first app

Steps 2 to 5 do not depend on each other. Step 6 waits on all of them; step 7 waits on step 2 (the extension cannot find your tenant until the domain is verified).

With one engaged person who has every right, the whole thing takes about an hour. Across three teams with ticket queues, expect one to two weeks, almost all of it waiting.

For AI agents: llms.txt, llms-full.txt, or any page with a .md suffix.