On this page
Key facts at a glance
| Question | Answer |
|---|---|
| How many write-ups are here? | Five. That is the public flagship set. |
| Who keeps the HighLevel account? | You do. The sub-account and the data stay in your name. |
| Why not one giant app? | The first custom app pays for OAuth, install and uninstall, webhooks and data sync. The next products sit on that layer. |
| Messy account already? | Start with a written audit, not a build. |
| Which API? | Current work is HighLevel API v3: scoped OAuth, tokens and webhooks. |
| Is this a snapshot gallery? | No. These are products and server-side jobs, not a snapshot import. |
What HighLevel case studies are on this site?
There are five public write-ups: the provisioning console, FrontDesk Assist, the Zapier to n8n migration, automated client onboarding through the API, and the platform-layer story that ties the two apps together.
They are anonymised where they need to be. None are a composite cover pretending to be a product.
Product screens and diagrams live on the write-ups: FrontDesk queue shots, the console source-tier legend, the account-setup schematic on onboarding, and the orchestration diagrams on the n8n page.
What is the provisioning console?
The provisioning console takes a new HighLevel sub-account from blank to a reviewable draft in minutes, instead of someone typing about a hundred fields by hand.
The build is identical every time. The content inside it is different every time. Each proposed value has a source and a confidence figure, and a human reviews the draft. Read the provisioning console write-up. This is the Marketplace-app class of work.
What is FrontDesk Assist?
FrontDesk Assist is a Marketplace-shaped app: live queues, a speed-to-lead clock and a team leaderboard, installed into the HighLevel sub-account so leads do not die at the desk.
You can deploy a clean CRM and still watch the floor fail to own the next call. It white-labels for agency resale. Read the FrontDesk Assist write-up. Same class of work as other custom Marketplace apps.
What happened when Zapier pricing broke at volume?
Per-task Zapier pricing stopped making sense at volume, so the automation moved onto self-hosted n8n: lower cost per run, and headroom to scale.
Native connectors and Zapier are the right first move. At volume the bill scales with success and the failure modes live in someone else's queue. Read the Zapier to n8n write-up. That sits with self-hosted n8n and the MCP and API class of work.
What does automated client onboarding through the API look like?
Automated client onboarding is server-side setup of a new HighLevel account through the API, instead of a person working down a checklist.
That is HighLevel API v3, not a dashboard ritual. Read the automated onboarding write-up. See MCP servers and APIs. Messy accounts still start with a written look.
Why do two apps sit on one platform layer?
Two apps sit on one platform layer because the first custom app pays for OAuth, token handling, install and uninstall, webhooks and data sync, and the second does not get that work rewritten.
That is why two apps exist rather than one, and why the second took a fraction of the time. Read the platform layer, then the provisioning console and FrontDesk Assist write-ups. The product surface is AgencyPRO.tools.
The same pattern applies beyond agency apps, for example the donor management system for nonprofits, built on HighLevel custom objects and a logged sync. Donor is a related product, not a sixth case study.
How should you use these if you are hiring?
Read the write-up that is closest to your job, then start with a written audit if the account is messy, or a scoping call if you already know you need this class of build.
- Pick the closest case above.
- If the account is already noisy, request the free account audit (or use the form below).
- If you need a developer for the platform layer, hire a GoHighLevel developer.
- How the work runs: engagement.
Two useful notes first: what a GoHighLevel developer actually does, and signs your GoHighLevel account needs an audit.
FAQ
Can you name the client?
Usually no on this site. Public write-ups are anonymised so the architecture is visible without putting a client's brand on a marketing page. AgencyPRO.tools product surfaces can be named because they are Gareth-associated brands.
Will you work inside my HighLevel account?
Yes. You grant access. You keep the login and the data. Nothing is moved onto a private instance you do not control.
Is a case study the same as a snapshot?
No. A snapshot copies dashboard configuration into another sub-account. These write-ups are products and server-side jobs: Marketplace-shaped apps, API provisioning, and self-hosted automation.
When is this a Marketplace app versus a one-off script?
A Marketplace app installs through HighLevel, with OAuth and a life after the first import. A one-off script is a job for one account. FrontDesk Assist and the provisioning console are the app pattern. Automated onboarding through the API can be either, depending on whether other locations will install it.
Do I start with a build or an audit?
If the account is already messy, start with a written audit. If you already know you need the same class of build as one of these five, book a scoping call and say which write-up is closest.
Next step
If one of these is the job, say so
Thirty minutes is usually enough to tell whether you need this class of build, or a written audit first. You keep the account either way.
Not sure yet? Start with the free audit.
Gareth Richardson
HighLevel Verified Developer Partner and Certified Admin.
I audit and rescue HighLevel accounts, then build the slices the menus cannot do: API v3 integrations, Marketplace apps, and Voice AI that writes into the same CRM. You keep the sub-account and the data throughout. More on the about page.