Pricing
Every plan is
the whole product.
Three plans, and no feature behind any of them. What changes is how much of it you use — how many people, how much you keep. Start free, with no seat minimum and no credit card.
The plans
They differ by size — how many people, how much you keep. Never by what the product can do.
Save up to 20% paying yearly
| Limit | Free $0 forever Five people or fewer, running for real. Start free | Growth $20 per member / month, billed yearly You passed five people or three projects. Try free for 14 days | Enterprise Custom annual Procurement, security review or custom terms. Talk to us |
|---|---|---|---|
| People | 5, in any mix | As many as you need | Unlimited |
| Read-only viewers | Included in the 5 | $5 each | Negotiated |
| Projects | 3 | Unlimited | Unlimited |
| Storage | 2 GB | 25 GB + 5 GB per member | Negotiated |
| Largest file | 250 MB | 1 GB | Negotiated |
| History | 90 days | 2 years | Unlimited |
| AI credits | 50 / month | 250 per member / month | Negotiated |
| CI results | 5,000 / month | Unlimited | Unlimited |
Free is not a trial, a sandbox, or a demo that expires — it is the whole product for a team small enough that we would rather have you using it. Prices are per member; read-only viewers, who read results and never write a test case, are $5 on Growth where most tools charge them full rate. Free does not price the difference at all — its five seats are five people, in whatever mix suits you.
For example: ten members and twenty read-only viewers on Growth is $300 a month — for thirty people, all of whom get the whole product. Storage and AI scale with the ten members rather than the thirty people, since a read-only seat costs a quarter and carries neither.
We don't hide features.
We size the plan.
The usual way to build tiers is to take the product apart and put the good half behind a higher price. It works, and it means the tool you evaluated is not the tool you end up with.
So we're doing the other thing. Every plan is the whole product. What changes is how much of it you use — and if you're a three-person team, the answer is that you use all of it for nothing, indefinitely.
Enterprise is the one place that needs a footnote, so here it is: it adds procurement and service terms around the product — a named contact, a security review, custom terms and invoicing — and nothing inside it. Every capability in the table below is on all three plans.
Everything, compared
Every row below is on every plan, including the free one. That is what the table is for — the ticks are the argument, not the decoration. Only the last two groups differ.
| Feature | Free | Growth | Enterprise |
|---|---|---|---|
| Test management | |||
| Test cases, folders & suites | ✓ | ✓ | ✓ |
| Shared steps | ✓ | ✓ | ✓ |
| Parameters & datasets | ✓ | ✓ | ✓ |
| Custom fields | ✓ | ✓ | ✓ |
| Bulk operations & filtering | ✓ | ✓ | ✓ |
| Excel, CSV & JSON import / export | ✓ | ✓ | ✓ |
| File library with previews | ✓ | ✓ | ✓ |
| Requirements & traceability | |||
| Requirements with hierarchy | ✓ | ✓ | ✓ |
| Tasks | ✓ | ✓ | ✓ |
| Test-case linking | ✓ | ✓ | ✓ |
| Coverage roll-up | ✓ | ✓ | ✓ |
| Execution & releases | |||
| Manual test runs | ✓ | ✓ | ✓ |
| Release-gated executions | ✓ | ✓ | ✓ |
| Go / No-Go readiness scorecard | ✓ | ✓ | ✓ |
| Requirement coverage & burndown | ✓ | ✓ | ✓ |
| Frozen sign-off snapshot | ✓ | ✓ | ✓ |
| Evidence & notes on every step | ✓ | ✓ | ✓ |
| Defects | |||
| Native defect tracking, tied to the failing step | ✓ | ✓ | ✓ |
| Kanban board with swimlanes & WIP limits | ✓ | ✓ | ✓ |
| Cumulative flow & cycle time | ✓ | ✓ | ✓ |
| Aging, MTTR & reopen rate | ✓ | ✓ | ✓ |
| Duplicate detection | ✓ | ✓ | ✓ |
| Oracle — the AI in Hawzu | |||
| Test case generation from specs, PDFs & Word | ✓ | ✓ | ✓ |
| Existing-coverage judge | ✓ | ✓ | ✓ |
| Documentation gaps named, never invented | ✓ | ✓ | ✓ |
| Canon grounding with citations | ✓ | ✓ | ✓ |
| Atlas map proposals | ✓ | ✓ | ✓ |
| Rewrite — free text & whole test case | ✓ | ✓ | ✓ |
| Ask: records, charts, cited answers & forms | ✓ | ✓ | ✓ |
| AI release-readiness narrative | ✓ | ✓ | ✓ |
| Analytics & reports | |||
| Observatory question-first dashboards | ✓ | ✓ | ✓ |
| Custom chart builder | ✓ | ✓ | ✓ |
| Project health score | ✓ | ✓ | ✓ |
| Flaky-test detection | ✓ | ✓ | ✓ |
| Release readiness, traceability, coverage & automation-health reports | ✓ | ✓ | ✓ |
| Sign-off PDF export | ✓ | ✓ | ✓ |
| Automation & CI | |||
| Trigger Jenkins, GitHub Actions & GitLab CI | ✓ | ✓ | ✓ |
| Import JUnit, TRX, NUnit, Cucumber & Allure | ✓ | ✓ | ✓ |
| Results mapped to test cases by code | ✓ | ✓ | ✓ |
| Build reports archived as evidence | ✓ | ✓ | ✓ |
| Integrations & API | |||
| Jira, GitHub, GitLab, Jenkins & Azure DevOps | ✓ | ✓ | ✓ |
| Linear & Slack | ✓ | ✓ | ✓ |
| Signed webhooks | ✓ | ✓ | ✓ |
| Public REST API | ✓ | ✓ | ✓ |
| Scoped access tokens | ✓ | ✓ | ✓ |
| Read-only MCP server for AI clients | ✓ | ✓ | ✓ |
| Governance & access | |||
| Roles & permissions | ✓ | ✓ | ✓ |
| Custom roles | ✓ | ✓ | ✓ |
| Audit log | ✓ | ✓ | ✓ |
| Subscriptions & notifications | ✓ | ✓ | ✓ |
| Limits The same numbers as the table at the top of the page. | |||
| People | 5, in any mix | As many as you need | Unlimited |
| Read-only viewers | Included in the 5 | $5 each | Negotiated |
| Projects | 3 | Unlimited | Unlimited |
| Storage | 2 GB | 25 GB + 5 GB per member | Negotiated |
| Largest file | 250 MB | 1 GB | Negotiated |
| History | 90 days | 2 years | Unlimited |
| AI credits | 50 / month | 250 per member / month | Negotiated |
| CI results | 5,000 / month | Unlimited | Unlimited |
| Support & terms Procurement and service terms around the product, not features inside it. | |||
| Email support | ✓ | ✓ | ✓ |
| Named contact | — | — | ✓ |
| Security review | — | — | ✓ |
| Custom terms & invoicing | — | — | ✓ |
What we measure
Five things, and nothing else. Every plan has all the features; these are the only dials that move between them.
- People Everyone in the workspace. Read-only viewers — who read results and never write a test case — cost a quarter of a member on Growth, because reading is not the same job as writing. On Free there is no difference to price: five people, in whatever mix suits you.
- Projects How many products or teams you are running at once.
- Storage Screenshots, recordings, and the CI reports archived against a build. On Growth the per-member share is added for members, not for read-only viewers.
- History How far back the audit trail and the trend charts reach. Your test cases, requirements and releases are never on a clock — only the derived history behind them.
- AI How much generation, Oracle and grounding you use in a month — about 600 written test cases on Free, about 3,000 per member on Growth, pooled across the workspace. Free's allowance is the same whether you are one person or five. The only meter that costs us real money per use, and the only reason any of this is metered at all.
Questions
The ones a pricing page actually gets asked — including the two whose honest answer is "no".
Is the free plan a trial?
No. It does not expire, it does not need a card, and it is not a cut-down build — five people get the whole product, including the AI, indefinitely. The five are people, not roles: fill them with writers, with read-only stakeholders, or any mix. The limits are about size, not capability. There is a separate 14-day Growth trial if you want to try the larger limits, and when it ends you land on the free plan rather than losing your work.
What happens when we outgrow Free?
Start the 14-day Growth trial — no card, and it begins the moment you sign up. It runs once per workspace, and a workspace that has already been on a paid plan cannot start another. After that you switch to Growth yourself in workspace settings: pick monthly or yearly, choose your seats, and the dialog prices the first period before you confirm. We invoice you for it — there is no card form yet, which we would rather say plainly than hide behind a checkout that does not exist.
What counts as a read-only viewer?
Someone who reads results and never writes a test case — the product manager checking a release, the engineer looking at a failure. They are $5 on Growth, a quarter of a member, because reading is not the same job as writing, and most tools bill them at full rate. Two things worth knowing: a read-only seat does not add storage or AI credits, since those scale with members; and on Free the distinction is not priced at all — five people, any mix.
How many test cases can the AI write for us?
About 600 a month on Free, and about 3,000 per member on Growth — so a team of ten has roughly 30,000, pooled, which is more than any team of ten will write. Those figures are measured from real requirement documents at the most expensive settings, so the usual number is higher.
What is an AI credit?
The meter behind that number. One credit is one cent of list-price model spend, and a written test case works out at about a twelfth of one — so 250 credits is roughly 3,000 cases. Growth includes 250 per member per month, pooled across the workspace, so the people who use AI heavily draw on the budget of those who do not.
Does connecting an AI client use our AI credits?
No. Credits pay for the model Hawzu runs on your behalf — writing test cases, or Oracle turning your question into a query. Over MCP that reasoning is done by the agent you connected, on your own account with whoever provides it, so Hawzu runs no model and meters nothing. The MCP server is on every plan, including the free one.
What is the difference between paying monthly and yearly?
The rate. Every price on this page is per member per month; yearly is the same seat at a lower monthly rate, charged once for the twelve months ahead. Monthly is charged each month and you can change or stop it whenever you like. Switching to yearly takes effect immediately and re-prices the month you are in down to the days you actually used; switching back to monthly waits until the year you have paid for is over, so nothing you have already bought is lost.
What if someone joins or leaves in the middle of a month?
A seat added part-way through is charged only for the days it is there, so nobody pays a full month for someone who arrived on the 28th. A seat given up stays on the bill for the period it was in and is simply absent from the next one — we do not refund part of a period, and we do not charge twice for the same person moving between seat types. Nothing about your team size is guessed: the invoice is built from the seats the workspace actually held.
What happens if we run out of AI credits?
New AI requests wait until the month resets, and everything else in Hawzu keeps working exactly as before. You are never billed for going over — we would rather pause than send a surprise invoice, which is the complaint most usage-priced tools generate.
Does the history limit delete our data?
No. Your test cases, requirements, releases and defects are never on a clock, on any plan. The window applies only to derived history — the audit trail, the change feed and the analytics behind trend charts — and it hides rather than deletes, so a longer window gives you your past back instead of finding a hole where it used to be.
Do you support SSO, SCIM or self-hosting?
Not today. Hawzu is cloud-only, and single sign-on is on the roadmap rather than in the product — we would rather tell you now than at the end of a procurement review. If one of those is a hard requirement, say so and it will weigh on what gets built next.
Can we pay in Indian rupees?
Yes. Growth is ₹1,499 per member per month billed yearly, or ₹1,899 monthly, with read-only viewers at ₹375. That is a price set for the Indian market rather than converted from the dollar one.