Intern
What you actually do
Permissions test and the checklist. Change log row. You do not 'customize' a location pipeline.
Library / Wire / Locations will edit sacred pipelines unless permissions and a change log exist
Franchise and multi-location GHL dies in permissions: every site rewrites stages, tags, and automations, then HQ reporting is a negotiation. Wire what is locked, what is local, who can publish workflows, and how a change is logged. This is the operating layer for snapshots — not a click-path through Agency settings.
Franchise and multi-location GHL dies in permissions: every site rewrites stages, tags, and automations, then HQ reporting is a negotiation. Wire what is locked, what is local, who can publish workflows, and how a change is logged. This is the operating layer for snapshots — not a click-path through Agency settings.
Write locked vs local before the next snapshot. If everything is editable, everything will drift. If nothing is local, locations will work in WhatsApp.
Roles: HQ admin, location manager, front desk. Front desk should not publish agency-wide workflows. That sentence saves you a quarter.
A change log is a page: date, what changed, breaking or safe, who was notified. Without it, every update feels like an attack.
Pilot one location. Do not email a snapshot to thirty sites on a Friday. Training is inside the rollout, not a Zoom after they are already live on broken tags.
Stolen leads and cross-location visibility: routing rules in writing. If locations can see each other’s contacts without agreement, you have a culture incident waiting.
Intern
Permissions test and the checklist. Change log row. You do not 'customize' a location pipeline.
Operator
Governance is the product. Pilot, version, notify, rollback. Local exceptions named. HQ definitions sacred.
CEO
Multi-location without permissions is a zip file. Locked definitions are how you see the business. Until they exist, more locations multiply folklore.
Agency account, sub-account/location, roles (HQ admin, location admin, user), locked pipelines/tags, local fields, snapshot version, change log, pilot location, rollback.
HQ locks pipelines, spec tags, core workflows; location cannot rename stages
Local exceptions named (offer names, hours, GBP link, user calendars) as local fields
Snapshot push → version note + who applied + QA checklist; not a Slack zip
User permissions: location users cannot export the whole account or edit sacred workflows
One page: pipelines, tag prefixes, core workflows, forms = locked. Calendars, GBP links, local offer copy, users = local. Anything else is a request to HQ.
Why: Unnamed exceptions become sacred in one site and invisible to HQ.
Done when: Locations can point to the page. Arguments go to the page, not to folklore.
GHL roles: who edits workflows, who exports, who deletes contacts, who can change pipelines. Remove 'admin for everyone' as a habit.
Why: Everyone-admin is how sacred pieces die over a weekend.
Done when: A location test user cannot rename a locked pipeline stage.
Version number, breaking vs safe, QA checklist (form write-in, calendar, missed call if relevant). Apply to a pilot first.
Why: Unversioned zips are how you cannot rollback and cannot train.
Done when: Pilot applied. Checklist signed. Then a second location, not thirty.
After each HQ change: log + message to locations with what they must not touch. Safe changes still get a note so silence is not a surprise.
Why: Surprise updates train locations to hide work in WhatsApp.
Done when: Last change has a log row and a notification. Rollback note if breaking.
Speed-to-lead, show rate, write-in — definitions locked. Locations cannot invent a parallel stage that HQ reporting misses.
Why: If stages drift, HQ dashboards are fiction and you will manage by anecdote.
Done when: One definition sheet. A location using a nickname stage is a ticket.
Print locked vs local. Create a location-manager test user. Try to rename a stage (should fail). Apply snapshot to a pilot with the QA checklist. Start the change log with today’s date. You are not rebuilding funnels for each site.
A company-wide snapshot blast, more locations on GHL, or a custom location request that edits sacred stages. Growth on drift is reporting fiction.
Zip emailed. Everyone is admin. Tags renamed locally. HQ dashboard does not match. Stolen-lead stories. Next snapshot overwrites local hacks and trust dies. That is a file, not a system.
Location user cannot edit locked pipeline. Pilot checklist passes. Change log has a row. HQ numbers use locked definitions. Second location only after pilot.
They need a named local field or a request. Tweaking stages is how roll-up dies. Offer copy can be local. Jobs cannot.
That is not a franchise system. That is thirty companies. Core is shared. Local is listed.
Then your change process is slow — fix the SLA on requests. The answer is not everyone-admin.
This week
This week: locked vs local page, roles, one pilot snapshot with QA, change log started. No Friday blast to all sites.
A full calendar and an empty pipeline is two databases.
Read ↗A high score in a CSV is not a qualified opportunity.
Read ↗If the thank-you does not change the contact, the page was a brochure.
Read ↗If paid lives in Stripe and the opportunity is still open, reporting is theatre.
Read ↗Working notes
The minimum CRM, pipeline, and nurture spec before you add more pages. Request the file. It arrives by email — not a public dump, not a drip of slogans.
Email and WhatsApp stay open. A call is for installing the system — not for a tour of tools.