Free Test Management Platform: Where Free Runs Out

Your spreadsheet has stopped coping. Two people edited the same row last week, nobody can say which cases ran before the last release, and there is no budget approved for a tool. So you search for a free test management platform, and you get the same thing every time: a list of ten tools in a comparison table.

The lists are not wrong. They are just answering a different question from the one you have. What you actually need to know is which of these will still work in a year, and what it will cost you when it stops.

Why this is not another list of ten tools

I checked page one for this search on 26 August 2026. Every result was a roundup: top seven, top ten, top fifteen free test management tools.

Read one closely and you will find the useful sentence. It sits near the bottom, under the table, in smaller type. Many free plans restrict reporting, integrations, or the number of users.

That sentence is the whole decision. Everything above it is furniture.

So this post skips the table. It explains instead what a free tier is built to do, and how to find where yours stops.

The three kinds of free

"Free" in this category means three genuinely different bargains. They fail in different ways, so it matters which one you are signing up to.

A free tier of a commercial product

A company sells a paid product and gives away a limited version of it. The limit is usually seats, projects or storage.

This kind is honest in one specific way: the boundary is written on the pricing page, in numbers. You can read it before you start.

It is also the one that ends most abruptly. A five user limit is fine right up until you hire a sixth person.

A community edition of an open core project

The project is genuinely open source. The code is public and the licence is real. Some of what a growing team needs sits in a paid subscription above it.

This boundary is harder to see, because it is spread across a features page, a pricing page and a few blog posts. It is rarely a single number.

Genuinely free, with support as the paid thing

The whole product is open and there is no tier above it. If anyone charges you, they charge for help, hosting or a contract.

Nothing is held back. The cost turns up somewhere else entirely: in the work of running it, and in how modern it feels to use.

What each one is designed to run out of

A free tier is not charity and it is not a trap. It is a commercial decision with a shape, and the shape is predictable.

• A free commercial tier runs out of seats. It is built to work beautifully for four people, so that the fifth starts a conversation.
• A community edition runs out of features. It works until you need single sign on, an audit trail, multi-tenancy, or a version you can pin.
• A fully open project runs out of your patience. Nothing is withheld. You simply own the database, the upgrades and the Friday evening.

Knowing which one you are in tells you which question to ask:

• Free commercial tier. How many people will realistically need access in two years?
• Community edition. What appears on the features page that is missing from the free column?
• Fully open project. Who on your team actually wants to own a server?

Where one free tier actually stops

Kiwi TCMS makes a good worked example, and not because anything is wrong with it. It is a healthy, actively maintained project. Version 16.3 shipped on 24 August 2026, it is GPL-2.0, and it has been pulled over 500,000 times.

It is a good example because the project documents every boundary openly on its own site. Here is where the free Community Edition stops, checked on 25 August 2026:

• The container displays ads from EthicalAds inside the instance you are hosting yourself. Ad free images come with the Self Support subscription at 25 dollars a month, per their post of 20 August 2025.
• Builds are x86_64 only. If someone on your team has an Apple Silicon laptop, or you run on Graviton, the free image does not cover it.
• It is rolling release with no version tagged images, so you cannot pin a version. You take whatever landed upstream.
• Multi-tenancy sits in the Enterprise plan at 600 dollars a month, per their features page and their post of 18 February 2026.

Here is how that lands in practice. A team of six adopts the free container. Eighteen months later they have 1,400 cases and a year of run history inside it.

Then two things happen in the same month. A new starter is issued an Apple Silicon laptop and cannot run the image. An auditor asks which version was running in March, and there is no version tag to point at.

Neither of those is a bug, and neither is anybody being tricked. Both were documented all along. The phrase "free and open source" simply never mentioned them.

The other bargain

TestLink is the third kind. It is genuinely free with no paid tier above it, and it has held a lot of teams together for a long time.

Checked on 25 August 2026: GPL-2.0, written in PHP, around 1.6k GitHub stars. It is still maintained, whatever the listicles imply. Version 1.9.20 is the last of the 1.9.x line, and a 2.x line is in development.

Its cost is not a paywall. It is that the interface shows its age, and that the XML export keeps almost no run history. If you ever leave, the cases come with you and eleven months of results do not.

What free does not remove

Whichever kind you pick, if you host it yourself then the licence fee is the only thing that went away.

You still own a Postgres instance, its version and its disk. You still take backups, and you still have to restore one occasionally to prove they work. You still read release notes and run migrations. Somebody is still called when it stops responding on a Friday evening.

For a team that already runs infrastructure, that is a small cost and the control is worth it. For a team of three with nobody who wants the job, it is exactly why a paid hosted plan is cheaper than it looks.

The one question to ask before you adopt

Ask it in this form, because the vague version gets a vague answer.

What specifically happens at the point we outgrow this, and what does it cost us then?

You can usually answer it yourself in ten minutes. Open the features page and the pricing page side by side. Every capability that appears on one and not the other is a boundary. Write down the ones your team will hit within two years.

Then check three things the pricing page will not tell you:

• Which CPU architectures have published builds.
• Whether version tagged images exist, or only a rolling tag.
• Whether the running instance displays or fetches anything you did not put there.

Should you just keep the spreadsheet?

Sometimes, yes. A spreadsheet is free in every sense, and a small team can run one for years without embarrassment.

Two signals mean it has genuinely stopped working:

• More than one person needs to edit it at the same time, and you have started losing changes.
• Somebody asked what was tested before the last release, and the honest answer took an afternoon to assemble.

The second is the real trigger. A spreadsheet holds test cases perfectly well. What it does not hold is the record of what you ran, when, and what happened. We argued that at length in the record a release rests on.

When free is the right answer for good

Free is not a stepping stone for everyone. It is a fine permanent answer when three things are true:

• Your team size is stable and sits comfortably inside the free limits.
• Nobody is asking you for evidence the tool cannot produce.
• Someone is genuinely willing to own the operational side.

If all three hold, use the free tool and spend the budget where it matters more. If any of them looks shaky over the next two years, at least go in knowing which one will break first.

For the licence side of this, which is a separate question from price, we covered it in what open source actually has to mean here. If you are already on a free tool and thinking about moving, the migration plan post covers what to carry across.