What changed in HighLevel
Every product release, filtered so the ones that matter to a serious build are easy to find.
Recent HighLevel releases worth knowing about
Titles, dates and tags are HighLevel's, from the public changelog feed. The commentary is mine. Every card links back to the original release note.
HighLevel
OFFICIAL CHANGELOG Product releases from ideas.gohighlevel.com, with my read on each
{{ statusText }}
STEP 1
Schedule n8n cron, hourly
STEP 2
Fetch & reduce RSS in, title/date/tags out
STEP 3
Publish JSON on my own domain
STEP 4
Read This page, no CORS problem
Written up in full
Case studies, in depth
Case study
Provisioning: the merge is the engineering Fanning out across narrow sources, reconciling values that disagree, and why every field carries a confidence score.
Read more ›Case study
Speed to lead, made visible Why putting the clock on the wall does more for conversion than another automation.
Read more ›Case study
When per-task pricing stops working Auditing what the automation actually does before rebuilding it somewhere cheaper.
Read more ›Case study
Onboarding as a process, not an event Automating the repeatable parts and leaving judgement where it belongs.
Read more ›Not sure whether a release affects your build?
Foundations
Setting up a HighLevel account properly
Nothing here is a snapshot. This is the configuration underneath one — the decisions that are cheap to make on day one and expensive to unpick six months in, once workflows, templates and reports all reference them. Most accounts I am asked to rescue are not broken because of what was built. They are broken because of what was skipped before building started.
Business profile and timezone
Legal entity name, trading name, registered address, contact details and — critically — the timezone. The registered name is what has to appear at the foot of a marketing email, and it is regularly not the name on the website.
A2P 10DLC brand and campaign registration
Submit on day one. Use the legal entity and EIN or company number exactly as registered, because a mismatch is the usual rejection. Register the campaign use case honestly — a mis-declared campaign is what gets numbers banned later.
Domains and email authentication
A dedicated sending domain with SPF, DKIM and DMARC, and separate subdomains for links and funnels so a funnel problem cannot damage email reputation. Warm the domain before volume. Reputation is earned once and lost quickly.
User roles and permissions
Decide who can see billing, edit workflows and export contacts before the first invitation goes out. Retrofitting restrictions onto people who already have habits is a conversation, not a settings change.
The stack
Tools I actually use, and why
Everything listed here is in production on real client work. Where a tool has a referral programme I use the referral link and say so — it costs you nothing and changes nothing about whether the tool is on this list.
Links marked REFERRAL earn me a commission if you sign up. Nothing is on this list because of that — every tool here is one I pay for and run on client work, and the unpaid ones sit alongside the paid ones without distinction.
Supabase
APACHE 2.0 Postgres with auth, storage and realtime on top. This is the database foundation under a custom marketplace app — per-location tokens, sync state and anything the CRM has no field for. Row-level security maps cleanly onto per-location scoping.
Read more ›Coolify
APACHE 2.0 A self-hosted PaaS, so everything else on this list deploys onto your own droplet with TLS and rollbacks handled. The difference between running open source and maintaining it by hand at two in the morning.
Read more ›NocoDB
AGPL 3.0 A spreadsheet interface over Postgres for the relational data custom objects cannot model — inventory, job sheets, multi-party records. Staff get a familiar grid; workflows get a proper API.
Read more ›Metabase
AGPL 3.0 Reporting across sub-accounts, which HighLevel dashboards cannot do — they stop at the account boundary. Point it at a warehouse fed from the API and an agency can finally see all its clients in one chart.
Read more ›Uptime Kuma
MIT Watches webhook endpoints, cron jobs and the sites you host. Integrations rarely fail loudly — they stop firing and nobody notices for a fortnight. A heartbeat check turns that into an alert the same hour.
Read more ›Redis
AGPL / BSD A queue in front of webhook ingestion. HighLevel will happily deliver a burst faster than your handler can process it; a queue absorbs the spike, retries the failures and keeps you inside rate limits on the way back out.
Read more ›MinIO
AGPL 3.0 S3-compatible object storage on your own disk. Useful once media volume is real — call recordings, client uploads, generated documents — without per-gigabyte egress charges or handing another vendor the files.
Read more ›Vaultwarden
AGPL 3.0 A Bitwarden-compatible vault you host yourself. If you hold credentials for dozens of client accounts, where those live is a question worth having a good answer to before somebody asks it in a security review.
Read more ›Start here
Start with the audit
A written architecture record and a staged plan, useful whether or not I build the rest.