What a 5-Person Team Really Pays for No-Code AI Agents in 2026: Zapier, Make, n8n, Lindy, Gumloop
For five people, Lindy costs 5 x $49.99 = $249.95/mo and pools 25,000 credits. Zapier Team starts from $69/mo for up to 25 users. n8n Pro is €50/mo billed annually with 10K executions and unlimited users. Gumloop Pro starts from $37/mo with unlimited seats and 20k credits. Make bills in credits, with a Free tier of 1,000 per month and no paid price in the fetched text.
This comparison is based on the vendors’ published pricing pages as fetched on 2026-11-12; we did not run these platforms hands-on. Every figure below is pinned to those pages, and where a page did not expose a number, the gap is stated rather than filled. The working rule for the rest of this post is simple: put fixed, predictable volume into deterministic workflows, and spend agent credits only where judgment varies. The five-seat math is where the surprises live, because per-seat billing, shared credit pools, and execution-based limits punish different mistakes. Lindy charges per active user, Zapier sells a flat Team plan for up to 25 users, n8n charges per execution with unlimited users, Gumloop removes seats as a cost lever, and Make counts credits. Each unit misreads differently, so compare them on the same five-seat question.
Who this is for (5-person team)
Picture five humans in one shared workspace: a founder, an ops lead, and three people who touch automations. You run a handful of stable workflows—sync a form to a table, move a record, send a status note—plus one or two agents for open-ended asks. The assumptions are fixed up front: five seats, prices fetched on 2026-11-12, and n8n kept in euros because that is how the page lists it. Nothing is converted, since a conversion would be a second source of error. Where a vendor bills per active user, all five seats count; where a vendor bills per execution or per credit pool, the five people share one allowance. Treat the arithmetic as copy-paste ready: multiply the listed per-seat price by five, or take the single plan that already includes your team, and compare included units rather than the sticker alone.
Cost model per platform
Each platform counts work differently, and the unit drives the bill more than the sticker does.
Zapier: tasks and agent activities, Team starting from $69/month
Zapier counts tasks, and a task is a successful action: “If the action isn’t successfully completed, it doesn’t count against your task limit,” and polling is free because “Zapier never charges you a task to check for new data” (Zapier pricing). Built-in tools cost nothing—Tables, Forms, Filter, Formatter, Path, Delay, Looping, Sub-Zap, Digest, and Storage are listed as not counting—but “Every step in a Zap and every external connector call uses tasks,” and “AI model tier, code runtime, and connector type each affect how many tasks an action uses” (Zapier pricing). Free is $0/month with 100 tasks and two-step Zaps, and Professional is “Starting from $19.99/month,” yet the page FAQ states “Free and Professional plans are for one user,” so five people need Team, “Starting from $69/month,” with “25 users” plus shared Zaps, folders, and app connections, SAML SSO, and priority support (Zapier pricing). Agents layer on separately: Free is “Forever” with “400 activities per month,” Agents Pro is “$33.33/month” (“$400 billed annually”) for “1,500 activities per month,” and on Team “your team shares a pool of Agents activities,” where activities cover behaviors, chat, browsing, and knowledge lookups (Zapier pricing). Task tiers run from 100 tasks/month up to 2M with a “Custom task limit,” but tier prices appear only in the interactive selector, so the fetched text states no dollar price per tier. For the step-level math, see Zapier AI step and task cost math.
Make: credits are the unit, and the fetched text shows no paid price
Make changed its billing unit to credits: “As of August 27, credits are the billing unit in Make,” with “most actions” consuming “1 credit” while some advanced features that use Make’s AI Provider “may use more” (Make pricing). Routers and the error handlers Rollback, Break, Resume, Commit, and Ignore count zero credits per the FAQ (Make pricing). The Free plan gives “1000 credits per month,” and when credits run out, “Your scenarios won’t continue to run, until credits are added again” (Make pricing). Extra credits come in bundles of 1,000 or 10,000 at fixed prices and expire after one month on a monthly subscription (Make pricing). Under 20 employees? The Start-up Plan offers “50% off Business” (Make pricing). One gap, stated plainly: the fetched page does not show Make’s paid-plan dollar prices, because the plan cards render dynamically, so no Make dollar figure appears in this post. The Zapier-side contrast lives in Make vs Zapier 2026.
n8n: executions priced in euros, Pro €50/mo billed annually
n8n’s unit is the execution, defined verbatim: “An execution is a single run of your entire workflow. It doesn’t matter how many steps are in the workflow or how much data it processes. It’s still a single execution” (n8n pricing). Steps and data volume do not change the count. Starter is €20/mo billed annually with 2.5K workflow executions, one shared project, five concurrent executions, and unlimited users; Pro is €50/mo billed annually with 10K workflow executions, three shared projects, up to 50 concurrent executions, and unlimited users; Business is €667/mo billed annually with 40K executions and is “Self-hosted only,” adding queue mode for scaling and Git version control (n8n pricing). Assistant credits run 1,600/month on Starter and “Up to 9,600/month” on Pro, while a self-hosted assistant is BYOK, so “no credit limits apply” (n8n pricing). The Community Edition is a standard, self-hosted version available on GitHub. The deeper tier walkthrough is in n8n pricing explained.
Lindy: per-seat credits, 5 x $49.99 = $249.95/mo into one pool
Lindy bills every active user: “from $49.99 a month for 5,000 credits each, and everyone draws from one shared pool” (Lindy pricing). For five seats the arithmetic is 5 x $49.99 = $249.95 per month before top-ups, and the pool is 5 x 5,000 = 25,000 credits per month. Top-ups are “anytime at $10 per 1,000 credits” (Lindy pricing). Credit size follows the ask: everyday asks run 2-250 credits, deep work 250-1,000, and big builds 1,000-2,500 (Lindy pricing). Monthly credits “don’t roll over,” and at the limit “Nothing breaks and there is no surprise bill. Lindy pauses credit-using actions until your credits reset,” while “Anything with outside impact waits for approval such as sending an email, updating a ticket” (Lindy pricing). The free tier offers “$50 in credits” that “last 7 days” (Lindy pricing). For the builder-versus-builder framing, see Lindy vs Gumloop vs Zapier agents.
Gumloop: credits plus an 8% orchestration fee, Pro from $37/mo
Gumloop’s Pro “Starts at $37 / month” with a “14-day free trial,” and the page lists an Orchestration Fee of 8% in its Pro column, with no further definition (Gumloop pricing). The same row lists “Included Credits 20k per month,” “Seats: Unlimited,” “Teams: Unlimited,” “Concurrent Workflow Runs 5,” “Concurrent Agent Chats 25,” “Models 35+,” and “Bring Your Own API Keys” (Gumloop pricing). Because seats are unlimited, five people sit under the single $37/mo starting price with one 20k-credit monthly allowance, five concurrent workflow runs, and 25 concurrent agent chats. The practical read is that headcount stops being a cost lever on this plan, while concurrency and credits become the constraints to watch.
Five-person monthly cost table
The table maps each platform onto the same five-seat question. “Listed price (fetched)” repeats the page, not a quote from this post.
| Platform | Billing unit | Plan for five | Listed price (fetched) | Included | Over-limit behavior |
|---|---|---|---|---|---|
| Zapier | Tasks; agents as pooled activities | Team (up to 25 users) | Starting from $69/month | Shared Zaps, folders, and app connections; Agents activities pooled; Free and Professional are one-user | Pay-per-task at a higher per-task rate, or Zaps pause if pay-per-task is off |
| Make | Credits | Not priced in fetched text | No paid price shown | Free 1,000 credits/mo; most actions 1 credit; routers and error handlers count zero | Scenarios stop until credits are added; bundles of 1,000 or 10,000 expire after one month |
| n8n | Executions (one run of the entire workflow) | Pro, unlimited users | €50/mo billed annually | 10K executions; 3 shared projects; up to 50 concurrent executions; assistant credits up to 9,600/mo | Not stated on the fetched page |
| Lindy | Per-seat credits, one shared pool | Team, 5 active users | 5 x $49.99 = $249.95/mo | 25,000 credits/mo shared (5 x 5,000); top-ups at $10 per 1,000 | Credit-using actions pause until reset; outside-impact actions wait for approval; no surprise bill |
| Gumloop | Credits plus orchestration fee | Pro | Starts at $37/month | 20k credits/mo; unlimited seats and teams; 5 concurrent workflow runs; 25 concurrent agent chats; 35+ models; BYO API keys | Not stated beyond the 8% orchestration-fee row in the Pro column |
Fetched 2026-11-12. n8n figures are in euros (€) and are not converted; Zapier, Lindy, and Gumloop are in dollars ($). “Starting from” and “Starts at” caveats come from the pages themselves, Make’s paid price is absent from the fetched text, and n8n’s €50/mo Pro figure is the annual-billing rate.
Where agents lose to workflows
Agents earn their keep on variable, judgment-shaped work, while fixed, high-volume, identical-every-time steps belong in deterministic workflows, because a workflow repeats the same step at the same unit cost every run and an agent’s burn moves with the model, the runtime, and the connector. Two vendor details make agent spend hard to budget. Lindy’s own size guide spans 2-250 credits for everyday asks, 250-1,000 for deep work, and 1,000-2,500 for big builds (Lindy pricing), so one ask can differ from another by orders of magnitude. Zapier states that “AI model tier, code runtime, and connector type each affect how many tasks an action uses” (Zapier pricing), so swapping a model can move the bill without touching the workflow. Arithmetic from the listed figures: at 1,000 credits per big build, Lindy’s shared 25,000-credit pool covers 25,000 / 1,000 = 25 such builds before pausing; at 250 credits per everyday ask, the same pool covers 25,000 / 250 = 100 asks. The full decision rule lives in the agent-or-workflow cost decision guide; the short version is to default to a workflow and promote a step to an agent only when the variation is real.
Lock-in and exit risk
Price is half the decision; the other half is how painful leaving is. Two vendors sell allowances that reset or expire at month-end. Make’s extra credits arrive in bundles of 1,000 or 10,000 and expire after one month on a monthly subscription (Make pricing). Lindy’s monthly credits “don’t roll over” (Lindy pricing), so an unused pool is use-it-or-lose-it spend. Top-ups are a separate purchase at $10 per 1,000 credits. The exit paths differ just as much. Among the five pages fetched, only n8n’s names a self-hosted version: “A standard, self-hosted version of n8n is available on GitHub” (n8n pricing; GitHub), the self-hosted assistant is BYOK with “no credit limits” (n8n pricing), and Business at €667/mo billed annually is “Self-hosted only” (n8n pricing). Gumloop includes “Bring Your Own API Keys” (Gumloop pricing), so model credentials are not trapped inside the tool. Zapier Team lists “shared Zaps and folders” and “shared app connections” (Zapier pricing) as team features. Before committing, answer three questions: where do the scenarios or workflows live, in the vendor’s cloud or on infrastructure you control; where do the credits live, in a pool that resets monthly or in a balance you can draw down; and who owns the keys, you or the platform.
FAQ
Do five people really need Zapier Team?
Team is the plan built for five seats. The FAQ states “Free and Professional plans are for one user” and “Team includes up to 25 users” (Zapier pricing). Free is $0/month with 100 tasks and two-step Zaps, and Professional is “Starting from $19.99/month,” but both are single-user (Zapier pricing). Five separate one-user subscriptions at the listed Professional starting price would be 5 x $19.99 = $99.95/month before any task-tier differences, while Team is “Starting from $69/month” for up to 25 users (Zapier pricing). The caveat is that the fetched text does not price each task tier, so compare the included task allowance alongside the seat math.
What happens when we hit our limit?
It depends on the unit. Zapier “switch[es] you to pay-per-task billing,” charged “at a higher per-task rate than your base subscription tasks,” and if pay-per-task is off, Zaps pause at the plan’s limit until the next usage period (Zapier pricing). Make stops outright: “Your scenarios won’t continue to run, until credits are added again” (Make pricing). Lindy is the gentlest: “Nothing breaks and there is no surprise bill. Lindy pauses credit-using actions until your credits reset,” with outside-impact actions waiting for approval (Lindy pricing). For n8n and Gumloop, the fetched pages do not state an over-limit behavior for executions or credits, so this post does not invent one.
Is self-hosting n8n the escape hatch?
That is the only self-hosted route named on the five pages. “A standard, self-hosted version of n8n is available on GitHub” (n8n pricing; GitHub repo), the self-hosted assistant is BYOK so “no credit limits apply” (n8n pricing), and Business at €667/mo billed annually is “Self-hosted only” and not offered on n8n Cloud (n8n pricing). What the fetched pages do not cover is the operating cost of running that server, so treat self-hosting as an ops commitment, not merely a price move.
Takeaway
Compare the five-seat arithmetic before the feature list. Lindy is 5 x $49.99 = $249.95/month for a shared 25,000-credit pool (Lindy pricing); Zapier needs Team from $69/month because Free and Professional are one-user (Zapier pricing); n8n Pro is €50/mo billed annually for 10K executions with unlimited users (n8n pricing); Gumloop Pro starts at $37/month with unlimited seats and 20k credits, plus an 8% orchestration fee in the Pro column (Gumloop pricing); and Make counts credits with a Free tier of 1,000 per month and no paid price in the fetched text (Make pricing). Keep the workflow-first default: deterministic steps for fixed volume, agents for judgment, and promote only when variation is real. Weigh exit risk next to price, because a pool that resets monthly or a centralized connection set is a cost you pay again on the way out. For the next step, read the agent-or-workflow cost decision guide, the Lindy vs Gumloop vs Zapier agent comparison, the Zapier AI step and task cost math, Make vs Zapier 2026, and n8n pricing explained.
📖 Related Reads
- NiteAgent — AI agent development, frameworks, and production patterns
- ToolBrain — tool reviews, LLM comparisons, and AI workflow guides
Cross-links automatically generated from NoCode Insider.