Your Agency Runs Itself.

No. Orchestror is entirely chat-driven. You type a request like "/gsc-opportunities" or "write a blog post about X" and your entities handle everything — API calls, research, writing, publishing. Zero coding required.

Each site gets its own isolated config — you enter the WordPress URL and an Application Password (generated from WP Admin in 30 seconds). The publishing entity can draft, preview, or publish directly. No plugins, no webhooks to set up.

Yes. Each site authenticates via OAuth with the minimum required Google scopes. Credentials are stored encrypted per-site and never shared across clients. Your data is used only to generate the analysis you request — it is never used to train models.

Every action that publishes, modifies, or sends something externally requires your explicit approval before it ships. Entities draft and propose — you review and approve. This Human-in-the-Loop gate means nothing goes live without you seeing it first.

Yes. Multi-site management is a core feature. Each site has its own isolated credentials, brand voice, entity configuration, and data. Switching between clients takes one click, and the proposal scales with your roster.

Most clients run their first GSC analysis and publish their first AI-assisted article within the first 24 hours. Measurable SEO improvement depends on your content cadence and site authority — in one documented case, a single entity-run content sweep moved a page from position 17 to 7.8 in one week.

Yes. The Schedules feature lets you set any entity to run on a schedule — weekly GSC reports every Monday, daily GBP post generation, bi-weekly content briefs. You get notified when they run and can review the output before anything publishes.

Less than you'd think. Data-heavy work — GSC pulls, crawls, backlink crunching — runs in plain code, and only the computed result reaches the model. The same job that costs a browser agent ~200,000 tokens costs an entity about 1,500 — that's what makes weekly sweeps across a full roster affordable.

Claude and Codex today, selectable per entity, per skill, and per workflow. Entities are defined by their instructions and client context — not the model underneath — so when the next great model ships, it plugs in as one more runner, with no migration.

One call. We walk through your roster, the channels you want the entities running, and how hands-on you want us — and you have a number typically within 48 hours. No feature matrix to decode: every proposal includes the full platform.

That's the design goal. Data pulls and number-crunching run in plain code — entities only spend tokens on reasoning — so a weekly sweep across 60 clients costs a fraction of what agent-style tools burn on a single account. Schedule it once and stop watching the meter.

There are no plans or tiers. Your proposal is shaped by three dimensions — how many clients you run, which channels the entities cover and how often, and how hands-on you want us. Everything is included at every size; you never pay to unlock a feature. One call, and you have a number within 48 hours. Your AI and data usage isn't part of the fee — Claude, OpenAI, and your data providers bill you directly, at cost, on your own accounts.

Yes — annual terms are available on any engagement and include a meaningful discount vs. monthly. It's part of the proposal conversation: hello@orchestror.com.

Yes — monthly engagements can be cancelled any time with no penalty, and your data and configurations are exportable on the way out. Annual engagements run to the end of their period.