Contract management shouldn't be a luxury purchase

August 2026

There are two contract management markets. One is served extremely well: large enterprises with a legal operations function, a dedicated budget line, and the appetite for an implementation project measured in quarters. The other is everyone else — the company with two hundred vendors, no legal department, and a procurement lead who is also doing three other jobs.

That second market is not badly served because it is difficult. It is badly served because it is inconvenient to sell to. So it gets a shared drive, a spreadsheet, and a calendar reminder that somebody set up two employees ago. We started Fratera because that gap is not a product problem. It is a pricing and packaging problem wearing a product costume.

What "affordable" usually means, and why it fails

The standard answer to a smaller budget is a smaller edition of the same software: take the enterprise product, remove features until the price works, and call the result an SME tier. It looks generous on a comparison table. It fails in a specific and predictable way.

You discover the missing piece at the worst possible moment. The repository is there, so you file everything — and then the renewal alert turns out to be an add-on. Approvals exist, but routing by delegation of authority is a tier up. E-signature is included, but only in the plan you did not buy. Each individual omission is defensible. Together they mean the tool cannot carry a complete process, so a parallel process grows up beside it, in email and spreadsheets, and now you are paying for software and doing the work manually.

A foundation that only holds part of the weight is not a cheap foundation. It is an expensive filing cabinet.

We started from procurement, not from legal

Most contract lifecycle management is built legal-first, because at enterprise scale the legal department is the buyer. That shapes everything downstream: the contract is the object, and the vendor is a text field on it.

For a mid-sized company the question is almost never "where is this clause?" It is "what are we actually committed to with this supplier, across every agreement, amendment and order form we have ever signed with them?" So we made the vendor a first-class part of the system rather than metadata — one view that adds up total commitment, shows what renews when, and tells you which agreements are load-bearing before you walk into a renegotiation.

That is a different product, not a cheaper one. It also happens to be the view that a procurement lead, a finance director and a managing director can all read without training.

What a foundation has to do to be worth building on

Our test for whether something belongs in the base product is simple: can a company run a contract from request to signature to renewal without leaving the system? If the answer is no, the thing we removed was not a premium feature — it was a floor tile.

So the foundation carries the whole path. A repository where the contract and its structured data live together. Renewal tracking that works from the date that actually matters, the notice deadline, rather than the renewal date everyone diarises by mistake. Approval routing that follows a real delegation of authority, so the right person signs off because of the value and the category, not because someone remembered to forward the email. Electronic signature in the same place the contract already lives. And the vendor view sitting over all of it.

We would rather ship one complete path than five partial ones. Bones first, then flesh — but never half a limb.

Three pricing decisions we made deliberately

Affordability is not a discount. It is a set of choices about what you are willing to charge for, and those choices are easier to judge than a headline number.

We do not count your contracts. Charging per contract is the most common model in this category and it quietly punishes the exact behaviour the product exists to encourage. If uploading your archive raises your bill, you will not upload your archive — and a contract repository with a partial archive is worse than useless, because it looks complete. Load fifty contracts or fifty thousand; the number is not our business.

You pay for the people who create and change contracts. Reading, approving and signing are free. An approver who opens a request once a month, a department head who signs twice a year, an auditor who needs to look something up — none of them should cost a licence, because the moment they do, someone starts sharing a login and your audit trail becomes fiction.

Limits are soft, and there are no surprise invoices. Where we do meter something, you can see the meter, and going past it starts a conversation rather than a charge or a lockout. Nothing in the product will stop a contract mid-workflow because a counter ran out. We think a supplier that can silently bill you more is a supplier you have to watch, and we would rather not be one.

Your data, in your own database

Every customer gets their own database rather than a shared table with a tenant column. It costs us more to run and it removes an entire class of failure in which one bug exposes one customer's contracts to another. Authorization is enforced in the database itself, not in the application, so a mistake in a screen cannot become a data leak.

You choose where that database lives — EU and US regions are live — and the answer stays true for the documents as well as the records. For a company whose contracts contain its entire commercial position, "where exactly is it, and who can reach it?" deserves a precise answer rather than a reassuring one.

AI where it helps, nowhere it decides

We use AI for the genuinely tedious part: reading an uploaded agreement and proposing the structured fields — parties, dates, values, notice periods — so a human can correct them in seconds instead of typing them in minutes. Everything it produces is a suggestion that a person confirms, and what gets stored is the confirmed result.

What it does not do is make decisions. No model determines who may see a contract, who may approve it, or what a commitment is worth. Those answers come from rules you configure and from the database that enforces them. An assistant that occasionally invents a number is a fine drafting aid and a catastrophic system of record, and we are building a system of record.

Who this is for

Companies large enough that contracts have become a real liability, and small enough that a six-figure CLM programme is not a serious conversation. If you know roughly how many suppliers you have, if your renewal dates live in more than one place, and if the honest answer to "what did we commit to?" currently involves opening a folder — you are who we are building for.

Contract management is infrastructure. Infrastructure should be boring, dependable, and priced so that having it is the obvious choice rather than a project you defend in a budget meeting. That is the whole intention.