Lindy AI's paid Team plan starts at $29.99 a month per user, with 3,000 credits for each seat, which works out at about one cent a credit. An everyday task such as drafting a reply uses somewhere between 2 and 250 credits, so 3,000 credits could cover anywhere from a dozen to over a thousand tasks.
That spread is the whole problem. Lindy publishes ranges, not a price per email, and the credits are shared across everyone in the workspace. The way to price it is to measure your own agents for a few days, then multiply, and to count seats as carefully as credits.
What the pricing page and the docs say, and where they differ
I read both pages on 5 November 2026, and they do not describe identical plans. That is worth knowing before you budget.
The pricing page lists a Free plan, a Team plan and an Enterprise plan. Free gives $50 of credits with no card, and those credits last 7 days. Team starts at $29.99 a month for 3,000 credits, and the page lists higher credit tiers in the same plan:
| Monthly credits | Price a month | Price per 1,000 credits |
|---|---|---|
| 3,000 | $29.99 | about $10.00 |
| 5,000 | $49.99 | about $10.00 |
| 8,000 | $79.99 | about $10.00 |
| 15,000 | $99.99 | about $6.67 |
| 35,000 | $199.99 | about $5.71 |
| 70,000 | $399.98 | about $5.71 |
The per-1,000 column is my own division, not Lindy's. It shows that the small tiers cost a flat cent a credit and the price per credit only falls once you reach 15,000. The page also lists tiers at 4,000, 6,000, 20,000, 45,000 and above; I have left them out because the point is the shape.
The docs page tells a slightly different story. It names three plans, Plus, Pro and Max, plus Enterprise, gives no prices, and sends you to Settings, then Billing, for them. It confirms the one-cent base rate: credits "cost $0.01 per credit on base plans". The docs also put computer use on Pro and Max and live onboarding on Pro and up, while the pricing page lists both under Team. My reading is that the two pages describe the same plans under different names, and the Team page is the one that shows numbers. Because I could not reconcile them, check your own Billing screen before you buy, and treat the table above as the pricing page's list prices that day.
The rules that apply to every plan
- Credits are used only while Lindy is working. An idle agent costs nothing in credits.
- Credits are pooled. Every seat adds to one shared pool, and admins can set a per-seat allocation so one enthusiastic user cannot drain it.
- Monthly credits reset each billing cycle and do not roll over.
- Top-ups cost $10 per 1,000 credits and stay available until used. Compare that with the 3,000-credit tier at about $10 per 1,000: a top-up is the same rate, without the commitment.
- When the pool is empty, credit-using actions pause until the reset. The docs say there are no surprise bills.
- Every active user in the workspace is billed as a seat, from $29.99 a month with 3,000 credits each, and each seat's credits join the shared pool. People who join the workspace or mention Lindy in Slack take a seat, though teammates who arrive through Slack get a 7-day trial first.
- Anything with outside impact, such as sending an email, updating a ticket or posting to another channel, waits for someone's approval. Read-only lookups from approved sources do not.
- If one task starts using far more credits than usual, Lindy pauses it and asks whether to continue or stop, so a runaway job does not quietly empty the pool.
- Upgrades take effect at once; downgrades wait until the end of the current term.
The Complete Excel Course with AI Integration
From formulas and pivot tables to Power Query, DAX, VBA and AI with Copilot, ChatGPT and Claude — hands-on from the first lesson.
By Sagnik Bhattacharya and Virginia Thorn, Primetime Emmy Award winner
Buy nowWhat a task burns: the published ranges
Lindy groups tasks by size, with credit ranges for each. These are the figures from the pricing page:
| Size of job (my label) | Credits | Examples the page gives |
|---|---|---|
| Everyday asks | 2 to 250 | Answers, summaries, lookups and updates in your tools, drafting a reply |
| Deep work | 250 to 1,000 | A competitor research report, triaging a day's support queue, turning a week of calls into a proposal |
| Big builds | 1,000 to 2,500 | Complex dashboards, decks built from CRM data, internal tools, heavy scheduled workflows |
At a cent a credit, an everyday ask costs between 2 cents and $2.50, deep work $2.50 to $10, and a big build $10 to $25. The docs add that Lindy picks a model per task to stretch credits and that you can choose the model yourself, which is one of the few levers on the bill.
What Lindy's usage guide adds
The docs page on usage rates jobs by credit impact rather than numbers, and it is a useful sorting tool. In the personal Lindy that each user gets, the daily brief, alerts on time-sensitive email, nudges when an ask goes unanswered and meeting prep briefings are all rated very low; catch-up summaries and meeting recaps are low. In the team Lindy that everyone @mentions, answering from approved sources and pulling a report are low, drafting a client update or document is moderate, team routines are moderate and priced by how often they run, and deep research varies. Lindy also ships built-in routines you switch on per connected account, including email drafting, email labelling, urgent email alerts, follow-up bumps and meeting scheduling, which map neatly onto most front-desk admin.
The practical reading: the always-on personal routines are cheap, and the credits go on drafting and research. So budget around how much drafting and digging you ask for, not around how many routines are switched on.
None of this says what your reply-drafting agent burns, because a reply that looks up a customer record, reads a long thread and writes three paragraphs sits at the top of the everyday range, while a one-line acknowledgement sits at the bottom. So the first job is measurement.
Worked example: a veterinary practice's front desk
Say a four-person veterinary practice has one person at the front desk who handles about 25 emails a day, books appointments and chases overdue vaccination reminders. The practice manager wants Lindy to triage the inbox, draft replies for the receptionist to approve, and prepare reminders. Only the receptionist will use it day to day, so the practice pays for one seat, starting on the 3,000-credit Team tier at $29.99. In Lindy's terms the jobs map onto the email labelling and email drafting routines, a daily brief, and a custom routine for reminders; because sending an email has outside impact, the reminders wait for the receptionist to approve them, which she does in one batch each morning.
The published ranges cannot tell her how many tasks 3,000 credits buys, so she runs a one-week trial on the free credits ($50 of credits is about 5,000 at a cent each, by my arithmetic) and reads the credit counter after each kind of job. Here is what a plausible log looks like. These numbers are illustrative, not Lindy's; yours will differ.
| Job | Per day | Credits each (measured) | Credits a day |
|---|---|---|---|
| Triage and label a new email | 25 | 6 | 150 |
| Draft a reply for approval | 12 | 40 | 480 |
| Vaccination reminder with a booking link | 6 | 15 | 90 |
| Daily summary of the day's bookings | 1 | 60 | 60 |
That is 780 credits a day. Over 22 working days it is 17,160 credits, almost six times the 3,000 in the base tier. At a cent a credit it would cost about $172 a month, so she is looking at the 15,000 tier at $99.99 plus a top-up, or a smaller agent. The honest answer to "how far does 3,000 go" for this practice is: about four working days.
What she trims
She looks for the jobs that burn credits without adding value. Triage is cheap per email, but 25 a day at 6 credits is the second-largest line, and most of those emails are newsletters. A rule in her mail app that sorts newsletters before Lindy sees them cuts the triage volume to about ten a day, saving 90 credits. Replies are the big line: a draft costs 40 credits. If only the ones that need a considered answer go to Lindy and the receptionist writes the routine confirmations herself, the 12 drafts become five. Now the day looks like this:
- Triage: 10 emails at 6 credits is 60.
- Drafts: 5 at 40 is 200.
- Reminders: 6 at 15 is 90.
- Summary: 1 at 60 is 60.
That is 410 credits a day, or about 9,000 over 22 days. The 8,000 tier at $79.99 plus a 1,000-credit top-up at $10 covers it, so roughly $90 a month for an agent that saves the receptionist perhaps an hour a day. Whether an hour a day is worth $90 is a sum only she can do, but now it is a real sum instead of "3,000 credits for $29.99".
Four more agents, priced the same way
An optician's appointment reminder agent
One job, run on a schedule: check tomorrow's eye tests and prepare a reminder with a rebooking link for someone to approve and send. Twenty a day, say 12 credits each, is 240 a day, around 5,300 a month. A single narrow agent like this is the best use of credits because the cost is predictable. The cost is also the first number to cut: reminders that a booking system already sends for free should not be paid for twice.
A pet shop's supplier chaser
The agent reads a weekly stock sheet, spots low items and drafts an order email to each supplier. Four suppliers a week, 80 credits a draft, about 320 credits a week or roughly 1,400 a month. Fits comfortably in the 3,000 tier, with room for something else. This is a "deep work" style job when the sheet is large; the cost rises with the size of the sheet it has to read.
A garden centre's quote follow-up
Ten new enquiries a week, each followed up twice. Drafting at 40 credits is 800 a week, 3,500 a month, and it needs a human's approval before anything is sent. This one scales with your marketing: a good spring weekend doubles the credits. Pooled credits help because the quiet autumn months leave room, but unused ones are lost at the reset, so do not size the plan for the peak.
A research agent that should not run on credits at all
A manager asks Lindy for a competitor price report every Monday. At deep-work rates of 250 to 1,000 credits, that is 1,000 to 4,000 a month, a third of a Team plan or more for something a person could do with a chatbot in fifteen minutes. A fair test of whether an agent is worth its credits is whether the same job costs less in a general assistant. The comparison is run properly in Lindy AI against Claude for admin tasks.
How to measure your own burn in a week
- Start on the free credits. The pricing page offers $50 in credits with no card, and the docs say free credits last 7 days. That is enough to run each agent a few times.
- Run each job ten times. Note the credits before and after, and write down the lowest, highest and typical. The spread tells you which jobs are predictable.
- Multiply by real volume. Count emails, bookings and enquiries from last month, not from guesswork.
- Add a third. Agents retry, long threads cost more, and an odd week happens.
- Pick the smallest tier above that number, and plan to top up in $10 steps rather than moving a whole tier.
The sums are the same ones shown in the worked example. If you prefer to see the trial approach applied to a real inbox and calendar, the free-trial test on calendar and email admin does it end to end.
A before-and-after for one email, with its credit cost
Credits follow effort, and effort follows how much the agent has to read and decide. The same enquiry can cost very different amounts depending on how the agent is set up. Take a customer asking a garden centre whether a delivery slot is free on Thursday.
Before (an agent told to "handle the inbox"): it reads the whole thread of nine earlier messages, searches the calendar for the week, looks up the customer, writes a four-paragraph reply with the full delivery policy and flags it for approval. Say that costs 120 credits, about $1.20.
After (an agent with a narrow instruction): "Read only the latest message. If it asks about a delivery slot, check Thursday and Friday in the calendar and reply in three sentences. Do not quote the policy." The same job now costs perhaps 35 credits, about 35 cents. The reply is shorter, and so is the approval, because the receptionist has less to proofread.
The difference, roughly three times, is the reason instructions are a pricing lever. A narrower brief, a shorter reply and fewer tools switched on all cut what the agent burns, and none of them costs anything to change. Both numbers are illustrative; the habit of comparing is the point.
A credit log you can copy
For the first month, keep a simple log, one line per agent per week, filled in on Friday. Here is a filled-in example for the pet shop's supplier chaser and a reminder agent:
| Week | Agent | Runs | Credits used | Per run | Note |
|---|---|---|---|---|---|
| 1 | Supplier chaser | 4 | 340 | 85 | One supplier sheet was huge |
| 1 | Reminders | 30 | 420 | 14 | Fine |
| 2 | Supplier chaser | 4 | 290 | 72 | After trimming the sheet |
| 2 | Reminders | 34 | 470 | 14 | Fine |
By the end of week two the owner knows that the chaser is the unpredictable one, because the sheet's size drives its cost, and that reminders are boringly steady. That tells her which to watch and which to forget about. If the chaser's per-run figure climbs, she shortens the sheet; if reminders climb, something has changed, such as the agent looping.
Picking the tier from the log, not the brochure
With two weeks of figures the choice is arithmetic. Take the total credits used in a typical week, multiply by 4.4, add a third, and read off the smallest tier that covers it. For the pet shop, weekly use of about 760 credits becomes 3,344 a month, plus a third, which is about 4,460. The 3,000 tier is too small and the 5,000 tier at $49.99 fits. If they had instead stayed on 3,000 and topped up the remaining 1,460 credits at $10 per 1,000, the cost would be about $29.99 plus $14.60, roughly $45 against $50, and any top-up credits left over would carry into the next month, which plan credits do not. The catch is that a top-up is a manual purchase each time, so a business that needs the extra every month may prefer the tier for the sake of not forgetting.
The decision changes at the top end. Beyond 15,000 credits the price per credit drops on the pricing page, so a business that really does use that much is better on the larger tier than on base tiers plus top-ups. Below it, top-ups at the same cent-per-credit rate give you flexibility without penalty.
Seats, caps and the credit-drain mistake
Because every active user is billed, a five-person team on the base tier is five seats at $29.99, about $150 a month for 15,000 pooled credits; the pricing page and the docs both describe the price as per user. Compare that with one person on the 15,000-credit tier at $99.99, the same credits for two thirds of the money. Neither page says whether every seat in a workspace must sit on the same credit tier, so read the checkout or ask support before planning a mix. If one person can run the agents and others only need the results, fewer seats is cheaper.
When one enthusiastic colleague empties the pool
Picture a practice manager who gives three colleagues access, each of whom is now a billed seat. In week one a nurse asks Lindy to summarise a long clinical-supplies catalogue several times, then to compare suppliers. That is deep work at 250 to 1,000 credits each time, and she presses continue when Lindy flags the heavy task. By the 12th of the month the pool is empty and the reminder routine has stopped. Lindy did say so, in a message nobody was watching, and the gap only came to light when a client rang. The two fixes are the ones the docs offer: set a per-seat allocation so the nurse has a share rather than the whole pool, and keep the customer-facing agents on a protected portion.
The other safeguard is not technical. Decide which agents are essential and write a manual fallback for each, one line: "if reminders stop, the receptionist sends them from the booking system". Pausing at zero is gentler than a surprise bill, but it still has to be someone's job to notice.
Deciding between Lindy and a cheaper rule-based tool
An agent earns its credits when the job needs judgement: reading a message, deciding what it is about and drafting something sensible. If the job is "when a form is submitted, add a row and send a template", a rule-based automation does it for far less and never runs out of anything. Lindy or Zapier for follow-up automation puts the two side by side, and Lindy against Gumloop does the same for another agent builder.
My own rule of thumb is to give Lindy the jobs where an unreliable human step is currently the bottleneck, and give rules the rest. If you would like help sorting your admin into those two piles, that is exactly what an AI implementation consultation covers, and the planning side is in how to prepare your small business for AI agents.
The sums, on one card
- Credits a month = jobs a day times credits per job times working days, plus a third for retries.
- Money a month = credits times about one cent on the base tiers, or the tier price for the larger ones; every user who touches Lindy is a seat from $29.99.
- Top-ups: $10 per 1,000 credits, kept until used. Plan credits are not.
- Before you buy: read your own Billing screen, because the pricing page and docs name plans differently.
Questions about Lindy's plans and credits
Why do Lindy's pricing page and docs name different plans?
When I checked on 5 November 2026, the pricing page showed Free, Team and Enterprise with credit tiers, while the docs referred to Plus, Pro and Max. Plans are being renamed or reorganised. Trust Settings, then Billing, in your own workspace, and ask support if the two still disagree.
Do unused credits roll over?
No. Plan credits refresh every billing cycle and do not roll over. Credits you buy as a top-up are different: they stay available until used. So buy a top-up for a one-off spike rather than raising the plan.
What happens when credits run out?
Lindy pauses credit-using actions until the next reset or a top-up, and the docs say there are no surprise bills. For a customer-facing agent, that means replies could stop mid-week, so set an alert and keep a manual fallback.
Does every teammate cost a full seat?
Per the pricing page, every active user in the workspace is billed, including someone who joins or mentions Lindy in Slack. Credits are pooled across seats, and admins can cap each person. Add only the people who really need the agent.
Further reads
- How to Set Up Human Review for AI Work Without Slowing Down — Keep a human check without slowing the work.
- Loom Business + AI: Is the AI Plan Worth the Extra Per User? — Loom Business at $18 against Business + AI at $24 per user: which AI features justify the $6, and a break-even sum for a small team.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: lindy.ai pricing page (plans, credit tiers, seats, credit usage and top-ups) and docs.lindy.ai Pricing, Billing, Usage and Limits and Routines pages, read on 5 November 2026.