Kyriaki Chaldaiou
Head of Procurement Strategy
Most businesses that tell me they have a contract register actually have a list of documents. Supplier name, what it is, where the PDF lives. It answers the question "do we have a contract with these people?" — which is almost never the question anyone is asking.
The questions people actually ask are different. Can we get out of this? Are they allowed to put the price up? Who decided to buy this, and do we still need it? When is the last day we can act?
A register built around documents cannot answer any of those. A register built around decisions can, and it does not need to be much bigger.
A contract register is a central record of what your business has agreed to — not where the paperwork is filed.
The distinction matters because the two lead to different columns. A document index records what the contract is. A register records what the contract does to you: what it costs, when it changes, what it takes to leave, and who inside the business is supposed to be paying attention.
For most SMEs the register starts as a spreadsheet, and for the first dozen contracts that is entirely sensible. The problems arrive later, and they are problems of maintenance rather than format.
Record the contracting entity, not the name on the invoice. Businesses routinely trade under one name and contract under another, and the difference surfaces at exactly the wrong moment: when you are trying to work out whether the company that just put its prices up is the company you signed with.
"Accounting software." "Office cleaning." "Equipment maintenance." The test is whether someone who joined last month can read the row and know what it covers. Internal project names and supplier product names both fail that test within about two years.
Someone inside the business who knows why the contract exists and whether it is still needed. Not the signatory, not the budget holder necessarily — the person who would notice if the service stopped.
Contracts without an owner are the ones that quietly renew for a decade. When people leave, their contracts usually stay, and nothing in a document index tells you that the person responsible for a given agreement left in 2024.
The start date matters more than people expect, because minimum terms, notice periods and price reviews are frequently calculated from it rather than from the anniversary you have in your head.
Record the term as it is actually written: 12 months, 36 months, rolling monthly, or ongoing until terminated. That last category is where the expensive surprises live — evergreen agreements that never reach an end date at all.
This is the field most registers either omit or record uselessly.
"90 days" in a column is information. It is not a date, and nobody acts on it. A contract ending 31 December with 90 days' notice has a real deadline of 2 October, and 2 October is the only date that should be in anyone's calendar.
If your register has one derived column, make it this one. The gap between the end date and the notice deadline is where most unwanted renewals happen, and it is entirely predictable from information you already hold.
Whether it expires, renews automatically, or becomes rolling — and if it renews, for how long. A twelve-month contract that auto-renews for twelve months is a very different commitment from one that rolls monthly, and the register should say which.
A number is not enough on its own. £12,000 might be the annual cost, the total committed spend, a minimum you have promised to reach, or an estimate based on usage that could go either way.
Minimum commitments in particular deserve their own flag, because they are the one figure that can cost you money even when you use less than you expected.
Some contracts let the supplier raise prices annually, linked to CPI, RPI, a fixed percentage, or simply their own list prices. Most businesses discover which one applies when the invoice arrives.
Recording it turns an annual surprise into something you can forecast — and occasionally something you can challenge.
Not every contract can be ended with notice. Some only allow exit at the end of the initial term; others have break clauses, termination for convenience, or conditions attached. A one-line summary saves an hour of reading at the moment you are least able to spare it.
Last, and genuinely last. Knowing where the document lives is useful. It is just not what the register is for, and registers that lead with it tend to stop there.
Full clause text. Scanned copies pasted into cells. Anything you would have to re-read the contract to maintain.
A register competes with the day job. Every field that takes judgement to fill in is a field that will be left blank by the third contract and wrong by the tenth. Keep the ones that are either copied directly from the agreement or calculated from something that is.
| Supplier | Service | Owner | Start | End | Notice | Act by | Annual | Renewal |
|---|---|---|---|---|---|---|---|---|
| ABC Software Ltd | Accounting software | Finance | 1 Apr 2026 | 31 Mar 2027 | 90 days | 31 Dec 2026 | £12,000 | Auto-renews 12m |
| XYZ Telecom Ltd | Business broadband | IT | 1 Jul 2025 | 30 Jun 2027 | 60 days | 1 May 2027 | £7,200 | Fixed term |
| Example Cleaning | Office cleaning | Operations | 1 Jan 2026 | Rolling | 30 days | — | £9,600 | Monthly rolling |
The bolded column is the one that does the work. Everything to its left is reference; that column is a decision with a date attached.
Note the third row. A rolling contract has no "act by" date because every month is a decision — which is either very flexible or completely invisible, depending on whether anyone has set a review date.
Do not try to capture everything at once. Find the contracts first — and the quickest route is usually not the filing system but the outgoing payments, because every live contract is attached to money leaving the bank.
Then fill in supplier, description, owner, start, term, notice period and value. Derive the act-by date. Everything else can be added as you touch each agreement for other reasons.
Businesses track what their system has a field for. Give a register a column for end dates and people will track end dates; give it a column for notice deadlines and they will start acting a quarter earlier.
That is the whole argument for building one deliberately rather than letting it accrete. A list of documents tells you what you own. A register tells you what you have agreed to, and how long you have left to change your mind.
Timemy keeps contracts and their key dates together, calculates the notice deadline from the terms rather than asking you to, and surfaces each one in advance. See how contract tracking works.
The short version: build the register around decisions rather than documents. Supplier, what it is, who owns it, when it started, how long the notice period is — and the date that notice period implies, which is the only field anyone will ever act on.
Timemy tracks your vendor contracts, calculates notice dates, and sends reminders before deadlines. Free for up to 10 contracts.
Get started free