What Helix AI costs

One ladder, three steps

Not a $500 look-over. Not a free workshop. A priced map with a number behind it.

Diagnose: $2,500, credited

The 72-hour diagnostic. One 90-minute recorded session with the founder and whoever touches a lead, then a readout three days later: the map of how work moves, the three things to build first, each one priced, and the number behind each. Credited in full to the build. You keep the map either way. If the readout does not map at least $10K of recoverable value in the business, full refund.

Build: scoped and priced in your readout

The second brain first, then the modules the readout picked, highest payback first. Every action drafts behind human approval from day one, and the second brain is typically live two weeks after kickoff.

Run: priced on the work

We run the system with you: monitoring, tuning, new modules as the company grows, usage checked at weeks 2, 4 and 8 from the logs, and a re-diagnosis every quarter. Documented handover, and you own the second brain and the code.

The second brain, in detail

A fixed-scope build that turns scattered company knowledge into one clean, AI-readable memory layer you own. It is the first thing we build, and it is priced in the readout with the rest of the build. No integrations to approve, no platform permissions to wait on, and every agent built later plugs into it. The monthly run keeps it current with a weekly curate cycle, a monthly re-test and a quarterly refresh, because the build is only 20–30% of a working system; the other 70–80% is the change management after it ships. The demo brain shows exactly what you get.

The monthly department

There is no rate card for the running department, because there is no standard department. It is scoped and priced in your readout, in writing, before anything gets built. The diagnostic decides which modules run, what they touch, and what they're worth against your own baseline; the monthly number comes out of that, and it doesn't move unless the scope does. If the honest answer is one module, that's what you'll be quoted. Nobody gets sold a bigger roster than their map supports.

Every module is built on the tools you already run, measured on your own numbers, and operated behind your approval. Capacity is real, not artificial: three build slots a month, because we build and run every system ourselves: that's the fractional AI department model. Departments go month-to-month after month six.

What moves the price

These are the drivers, and they're the same ones the diagnostic measures: how many modules you run; how many systems the agents touch (a CRM and an inbox is one thing; CRM, billing, community, ads, and a warehouse feed is another); approval complexity: how many action types, how many human checkpoints; and how much of your knowledge already exists in writing versus living in heads, which sets the size of the memory-layer work. The diagnostic weighs all four against your own numbers and turns them into one fixed monthly figure before you commit.

What's always included, and what isn't

• Included: validation of every agent action against its real side-effect; human approval gates until an action type earns autonomy; measurement on your own metrics, not a vendor dashboard; your data, dashboards, and runbooks stay yours; one-click token revocation; month-to-month after month 6.
• Not included: ad spend and your existing tool subscriptions. They stay in your accounts, where you can see them. Where AI usage runs on your own keys, that transparency is the point.

Where this sits in the market

The market runs from $297/month template tools (where you are the operator) to six-figure consulting builds with months of discovery. Helix sits in between, deliberately: systems we build and run, with the price published, a $2,500 diagnostic to start, and departments scoped to cost less per year than one mid-level ops hire. If a $297 tool genuinely fixes your bottleneck, buy the tool. The scorecard will tell you honestly which situation you're in.