BIZN.org
BIZN.org
The sticker price is not the job price. A "$0" plan with a 300-minute cap, a "$12" plan billed annually, and a plan billed per step can land in very different places once your actual monthly volume is counted. This page prices four jobs where the catalog's numbers support a real calculation — and says plainly where they do not. Everything runs in your browser; nothing is uploaded. Prices and limits below were checked on official pages on November 1, 2026 (catalog v1.7.0); verify on the official pricing page before budgeting.
The job, defined narrowly: turning spoken meetings into text, counted in transcription minutes per month. Otter.ai Basic (free) gives 300 transcription minutes per month as a shared pool with no rollover — and caps each conversation at 30 minutes. That second limit matters: a single 60-minute meeting does not fit the free tier even if your monthly total is under 300 minutes. Pro is $16.99/month billed monthly, or $8.33/month billed annually. Source record: EV-P-007 · Otter.ai in the directory →
Worked example, assumptions showing: 600 minutes a month. Free covers the first 300; the other 300 need Pro — $16.99 on monthly billing, or $8.33/month billed annually. If you also spend, say, 10 minutes reviewing each hour of transcript, that is 100 review minutes (600 ÷ 60 × 10); at $40/hour that adds $66.67, for a monthly total of $83.66 on monthly billing. Change the review numbers to your own — the example's 10 minutes is an illustration, not a measured average.
The job, defined narrowly: AI prompts — asking Grammarly's AI to rewrite, shorten, or improve text — counted per member per month. Free allows 100 AI prompts per month; Pro allows 2,000 per member per month at $12/member/month billed annually ($144 per member per year). The catalog lists no monthly-billed Pro price, so none is shown. Only Enterprise is unlimited. Source record: EV-P-008 · Grammarly in the directory →
Worked example, assumptions showing: one person, 250 prompts a month. Free covers 100; the other 150 need Pro — $12/month billed annually ($144/year). Beyond 2,000 prompts a month for one member, even Pro is exhausted and only Enterprise (custom price) is unlimited, so this calculator stops and says so rather than pricing a tier it cannot see.
The job, defined narrowly: workflow executions per month, where one full workflow run counts once however many steps it has. n8n Cloud Starter allows 2,500 executions a month at €20/month billed annually (US pages list $24/month); Pro allows 10,000 at €50 billed annually (US pages list $60/month). Pick the price list that matches the page you buy from — the two differ by region, and mixing them would corrupt the total. A workflow that polls every few minutes burns the allowance fast. Self-hosted Community edition software is free, but your server and your setup time are not in this total. Source record: EV-P-016 · n8n in the directory →
Worked example, assumptions showing: 3,000 executions a month on the European price list. Starter's 2,500 are exhausted, so Pro is needed — €50/month billed annually. The same 3,000 executions on the US list are $60/month billed annually. Beyond 10,000 executions, Pro is exhausted and Business/Enterprise pricing is custom, so this calculator stops there.
The job, defined narrowly: tasks per month, where every successful step costs one task and AI steps cost more. Zapier Free allows 100 tasks a month; Professional starts from $19.99/month. This scenario is deliberately thinner than the others, and that is the point: the catalog does not state Professional's task allowance, so once you pass 100 tasks this calculator can tell you the floor (at least $19.99/month) but not which paid tier your volume needs. A multi-step Zap burns the allowance several times faster than a one-step Zap, and with overage off, Zaps pause at the limit. Source record: EV-P-010 · Zapier in the directory →
Worked example, assumptions showing: 400 tasks a month. Free covers 100; the other 300 need a paid plan starting from $19.99/month — and because Professional's allowance is not in the catalog, the honest result stops at "at least $19.99/month; check Zapier's pricing page for the tier your volume needs." Compare scenario 3: the same workflow billed per run (n8n) versus per step (Zapier) is a different job price, which is why the billing unit matters more than the headline number.
These jobs are left out on purpose, because a number the vendor does not publish cannot be calculated without inventing it. If a vendor publishes it later, the scenario can be added; until then, "we cannot calculate this reliably" is the correct result.
| Job | Why it cannot be calculated from the catalog |
|---|---|
| Image generations (Adobe Firefly) | Free daily generations: the count is not published. Paid tiers run on generative credits that do not roll over, and video burns them fast — but the credit cost of your image or video is not a fixed published rate. Source: EV-P-009. |
| Agent research tasks (Manus) | Everything runs on credits, and the cost of a task is not shown before it runs. One deep-research task burning roughly 900 credits is an illustration from the catalog, not a rate you can budget from. Source: EV-P-021. |
| Automation runs (Make) | Free gives 1,000 credits a month, but one run can cost from 2 to thousands of credits depending on its modules — so credits cannot be converted into runs without knowing your workflow. Source: EV-P-015. |
| General chat (ChatGPT, Claude, Gemini) | Usage caps are expressed as rolling windows, high-demand throttling, or multiples of Free — not as monthly task counts. A "20 chats a day" input would be our invention, not the vendor's unit. Sources: EV-P-001, EV-P-002, EV-P-003. |
| Research from your documents (NotebookLM) | The vendor promises only "expanded access" per tier — exact notebook and source limits per tier are not published on the page checked. Source: EV-P-017. |
Take the subscription figure as the floor, not the forecast: it is the cheapest tier your stated volume fits, at the price checked November 1, 2026. Add your review time honestly — on real workflows it routinely costs more than the subscription, which is why the AI Model Cost Estimator is built around it, and why The Real Cost of an AI Agent counts setup and maintenance time too. If a result says a tier is exhausted or a number is unpublished, treat that as information, not as a failure of the page: it is telling you where the vendor's own disclosure stops. Nothing here is financial advice, and every figure should be verified on the official pricing page before you buy.
Browse the full directory → See the source behind every price → Estimate a model workflow → Find your first tool →