Pricing without seats

Binee is priced per workspace. Unlimited seats, every feature on every plan, and one meter: credits, spent only when Binee does work for you. The free plan includes 25 credits per 30 days, and subscriptions start at $12 a month. This post explains why we landed there, and is honest about the edges, including what happens when a workspace runs dry.

The problem with per-seat AI

Per-seat pricing made sense for tools where value scales with hands on keyboards. Everyone who edits documents needs a document seat. AI for business operations does not work that way, and pricing it per seat quietly breaks the product.

Run the math on a 12-person team. Four people would use an AI operations tool heavily. Three would use it weekly. Five would mostly benefit from what it posts into a shared channel. Under per-seat pricing you now face a bad menu: pay for 12 seats, most of them underused, or buy 4 and start rationing intelligence, deciding which teammates get to ask questions about their own work.

Rationing is the real damage. The whole value of a business brain is that it sees everything and everyone can query it. The account manager who was not worth a seat is exactly the person who needed the warning that a client asked for new work while their invoices sat unpaid. And Binee's team automations post findings into shared Slack channels where everyone in the channel sees them: a payment update or a client escalation lands in front of the whole team. Who exactly is the "seat" consuming that insight? Per-seat pricing cannot even describe how the product is used.

So we removed the question. One workspace, everyone in, each person connecting their own tools and getting their own feed. Adding your bookkeeper or your newest hire to Binee costs nothing, which means you actually do it, which means the brain sees more, which makes it better for everyone. Seat pricing punishes the exact behavior that makes the product work.

What a credit is

If seats are not the meter, work is. A credit is Binee doing work for you, and the rates are flat and inspectable rather than a token-count lottery:

• A chat question costs about 1 credit.
• Watching is free. The matching step that checks incoming activity against your automations costs nothing, no matter the volume. When something matches and Binee runs a full investigation, that firing costs a flat 1.3 credits.
• An automation's approved action step costs 1 credit per completed action, at most 3 per firing, so the most expensive possible firing of an acting automation is 4.3 credits. Only completed actions are charged, and the builder tells you this before you approve the step.
• Opening a dashboard is free, always, however often anyone opens it, and so is every filter, range, and shape change. A refresh, which recomputes the created cards and writes the Pulse, costs 0.5 credits; when its gathering is heavy, pulling several live tools or long histories, it is priced like the chat question it replaces. The exact charge always shows on the Billing page.

Note what is deliberately free: everything that scales with team size. More teammates reading dashboards, more people in the Slack channel where a team automation posts, more members asking to be added, none of that touches the meter. You pay when the brain thinks, not when someone looks at what it thought.

Model choices under the hood are our problem, not your pricing surprise: a credit costs what it costs regardless of which model did the thinking.

The controls, and the honest edges

Metered anything raises a fair fear: the surprise bill, or its quieter cousin, the silent overrun. Two mechanisms handle it.

First, each member has a daily budget: the number of paid automation firings Binee may run for them per day. It defaults to 50 and is adjustable from 1 to 500. Hit your budget and your insights pause for the rest of the day, resuming the next; your budget is yours alone, so a teammate's busy day never pauses yours.

Second, and this is the part most pricing pages skip: what happens when the workspace balance actually reaches zero. In Binee, paid work pauses honestly. New insight investigations wait until credits renew or are topped up; nothing is deleted, and everything resumes once the balance is back. If a firing's balance cannot cover its action step, the action is skipped and the insight says so, while the finding itself still arrives. Dashboard cards simply stop refreshing and show how old their numbers are, instead of posing as current. No degraded results dressed up as full ones, no silent stall you discover a week later. You can always ask Binee itself what your credits were spent on.

Every feature on every plan

The other decision worth defending: there is no feature matrix. Free and paid workspaces run the same product, automations, insights, dashboards, action steps, all of it. The plans differ in credits, not capabilities. We think gating features creates a dishonest evaluation: you trial a crippled version, guess whether the full one fits, and find out after paying. With 25 free credits per 30 days you run the real product on your real tools; the only question the upgrade answers is how much work you want Binee doing.

The summary is short. Seats meter people, and people are not the cost; thinking is. So we meter thinking, publish the rates, cap the daily downside, and pause honestly at zero. Bring the whole team, from $12 a month.