A few years ago I sat in a workshop where a team proudly walked me through their contract landscape. A situation I believe lots of people will recognize. Seven systems, each excellent at its own task. Then I asked a simple question: "Which of these tells me what we have actually agreed with this supplier today?" The room went quiet. Everyone had a piece of the answer, and nobody had the answer.

Innovation worth celebrating

I have worked in contract management and general leadership since the early 2000s, and I have never seen so much creativity as right now. AI review tools, clause analytics, obligation trackers, sourcing and intake platforms, e-signature: each solves a real problem, often brilliantly. Specialised applications are a sign of a maturing field, and I am glad to see them gain traction.

But there is a risk that deserves attention. Every new point solution is easy to justify on its own. Together, they can quietly turn a coherent process into a pile of tools, each with its own data, its own logic and its own version of events. Integrality, the ability to see and steer the whole contract lifecycle, is the first thing to go.

Three roles every landscape needs

In contract management and supply chain management, value comes from three distinct roles:

• A system of record: the place where the authoritative facts about a contract live, such as parties, terms, values, dates and status.
• A system of action: where work happens, such as drafting, negotiating, approving, monitoring and managing change.
• A repository: where the documents and evidences themselves are stored, versioned and retrievable.

In a simple landscape, one platform may cover all three. In a modern one, several tools share them, and that is where it gets complicated. When every system does its own thing in isolation, the roles blur. Two tools both believe they hold the truth about renewal dates. An obligation tracker is fed by a clause extraction nobody validated. A signed contract sits in a repository, while the record still says "in negotiation".

This is not an argument against buying

Let me be clear: I am not saying "don't buy separate systems." Best-of-breed tools can be exactly what your organisation needs. My argument is about sequence. Decide how the parts form a whole before you deploy them, not after the first audit or dispute exposes the gaps.

What holistic design looks like

From many implementations, I see the same elements in landscapes that work well:

A clear role for each system. For every tool, you can say in one sentence whether it records, acts or stores, and what it must never do.

Data ownership. Every data element has one owner and one place where it is created. Others consume it, they do not re-enter it.

One source of truth for contract data. People should never have to ask which system to believe. If two systems disagree, there is a known rule for which one wins.

Shared definitions and metadata. What counts as "active"? What is a "renewal date"? What is a "high-risk" clause? If every tool answers differently, reporting becomes an argument instead of a decision aid.

Integration and handovers. Data and documents must move between systems predictably, with clear triggers. Most failures I see are not inside a tool, but in the handover between two of them.

Governance. Someone decides when a new tool may join the landscape, how changes are made, and how data quality is monitored.

Process first. Design the end-to-end process, from need to renewal or exit, and only then place the tools on it. A tool should serve a step in the process, not define it.

An owner of the end-to-end outcome. Each tool usually has a business owner. Who owns the result across all of them? Without that person, nobody feels responsible for the gaps between the systems.

Seven questions before you buy the next tool

Before the next purchase, put these on the table:

1. Which role does this tool play: record, action or repository? Does it overlap with an existing one?
2. Which data does it create, and which does it need from elsewhere?
3. Where is the single source of truth for that data, and how will this tool respect it?
4. Does it use our definitions and metadata, or its own?
5. How does it hand over to the steps before and after it, and who maintains that connection?
6. Which governance rules apply to its data, access and changes?
7. Who owns the end-to-end outcome, and how will they see the whole picture?

If you cannot answer these questions yet, you are not ready to sign, and that is not a failure, it is the most valuable moment to find out.

A closing thought

Over the years I have learned that contracts rarely go wrong because a single task was done poorly. They go wrong in the spaces between tasks, teams and systems. Point solutions can make each task better. Only deliberate design makes the whole better.

So by all means, welcome the innovation. Just make sure your landscape remains something you can explain on one page, and that someone can answer the simple question I asked in that workshop: what have we actually agreed, today?

Merge all this into that other relevant question: "what do we want to achieve". 

By Arjen Van Berkum

Keywords: Legal and IP, Procurement, Supply Chain

Follow Us On

Become a Contributor Newsletter Signup