PRICING
Simple plans. No surprise charges.
Paid plans renew at the same price every month and include their full price as monthly AI compute — usage never triggers an automatic charge. Not ready for a plan? The free tier lets you browse the platform, keep portfolios, and run the quantitative pre-screens at no cost. Paid tiers include a 14-day first-time trial. Cancel anytime.
UPDATED August 20, 2026
| Plan | Monthly | Annual | Included compute | Alert screens | Fastest cadence | Alert channels |
|---|---|---|---|---|---|---|
| Free Browse the platform, keep portfolios, run free pre-screens | $0 | — | — (browsing + pre-screens) | 0 | — | — |
| Greenhorn For getting serious about a shortlist | $29/mo | $290/yr | $29 of compute | 3 | Daily | Email · in-app |
| Prospector The working-researcher tier | $69/mo | $690/yr | $69 of compute | 15 | Hourly | Email · in-app · SMS |
| Mother Lode Full cadence and webhook automation | $149/mo | $1,490/yr | $149 of compute | 50 | 15 min | Email · in-app · SMS · webhooks |
Annual billing is ten months' price for twelve — about 17% off. Stage 1 screening is free on every tier. Each column is defined below.
DEFINITIONS
What each column means
- Included compute
- Every paid plan includes its full monthly price as AI compute — the processing that powers deep dives and screens. A $69 plan includes $69 of compute each month, and the Account page shows how much is used and how much is left in real time, from the same meters that enforce the limits. If a month of research would run past the included compute, the extra draws only from a wallet you pre-fund yourself — usage never triggers an automatic charge. Without a subscription, the free tier covers browsing, portfolios, and the quantitative pre-screens; AI analyses come with a paid plan.
- Alert screen
- A saved screen RockTurners re-runs automatically in the background. When a company newly starts matching the screen’s criteria, it fires an alert; the same company on the same screen fires at most once an hour, so a match never floods a channel. The plan number is how many screens can be armed at once.
- Fastest cadence
- How often armed alert screens are allowed to re-check the market. On daily cadence a screen is evaluated once a day; hourly and 15-minute cadence shrink that window. In practice this is alert freshness: a daily-cadence match surfaces the next morning, a 15-minute-cadence match surfaces within the quarter hour.
- Alert channels
- Where a fire is delivered: an in-app notification, an email, an SMS text, or — on Mother Lode — a webhook. Every channel carries the same fire; higher tiers add more delivery paths.
Beyond the columns
- AI deep dive
- A second look at a single company with a stronger premium AI engine, run on demand from a scan result for extra conviction — the deeper verdict is stored beside the original for comparison. The AI re-reads that company's SEC filings — the 10-K, 10-Q, and whatever else the framework requires — against the framework's rubric and returns a cited verdict, with named unknowns wherever the filings are silent. A deep dive draws more of your included compute than a standard screen; ordinary screens and scans run on the standard engine and cost less. Fair-use limits apply and reset monthly.
- Webhook automation
- A webhook delivers each fire as structured data — an HTTP POST carrying the screen, the company, and the match details — to a URL you control, the moment it fires. That lets alerts flow into Slack, a spreadsheet, or an internal dashboard with no inbox in the loop. It is a notification pipe, not order routing: RockTurners never connects to a brokerage and never places a trade.
COST TRANSPARENCY
Every scan shows its cost before it runs
Independent of the plan, each scan displays its estimated processing cost up front and reports what it actually spent when it finishes. The Account page tracks the same numbers live — compute used and compute left — from the same meters that enforce the limits. Fair-use limits on premium deep dives reset monthly; nothing rolls into an automatic charge. The plan gates how much you can run — never what you can see: every subscriber gets the same verdicts, the same citations, and the same named unknowns from the frameworks they run.
Subscriptions are managed in the app and billed through Stripe. More questions answered on the FAQ — including what a verdict is and why the output is impersonal research, not advice.