“It’s Pretty Much Done” — Famous Last Words?

When someone says “It’s pretty much done”… what do you usually find?

Quite a lot, actually. Even when something looks fine at a glance, once you dig in, the cracks appear. I often see inconsistent terminology, mismatched documents, missing limits of liability, no exclusions for consequential loss, vague or unrealistic payment terms, or just terrible formatting that makes the whole thing hard to follow.

Here are my two biggest red flags:

1. No proper scope. It’s surprisingly common to get a purchase order or short-form contract that doesn't actually refer to the proposal, quote, or any detail about what's being delivered.
2. Foundational errors. For example, flawed payment terms that mean there’s no valid consideration and the contract may be unenforceable.

So when someone says it’s “ready to go,” I usually find that it’s not.

Somewhere in between. It has come a long way in the last couple of years, and it definitely leans toward helpful… but only if you know what you’re doing.

AI can be a great tool for idea generation or testing alternative ways to draft a clause. It’s like a sounding board available for a sanity check 24/7. But it doesn’t yet have the industry-specific knowledge or legal reasoning to draft an entire contract with confidence. Sometimes it jumbles legal concepts, gets the jurisdiction wrong or overlooks basic commercial realities. I also think people overlook the security and intellectual property issues when pumping sensitive information into an AI platform.

If you already know a lot about contracts, you can use AI to save time and explore different drafting options. But if you don’t, you’re likely to spend more time checking and fixing the output than you would just drafting it yourself. So at this stage, it’s not a replacement, but it can be a handy assistant.

 What’s something even experienced teams tend to miss?

Experienced teams tend to underestimate how stale their templates can get.

When you're using the same contract over and over again, it's easy to stop seeing its flaws. I call it "domestic (contract) blindness". It’s that sense of comfort that comes from something familiar, even when it’s no longer fit for purpose. Just because a clause hasn’t caused a problem doesn’t mean it’s working well.

I always encourage legal and contract management teams to periodically carve out time (outside of project delivery) to look at their contracts with fresh eyes. They need to ask themselves: Can this be clearer? Is this process realistic? Can we remove any friction? Regularly refreshing your templates helps avoid confusion when things go wrong, which they eventually will.

 You say a contract is only as strong as how it manages expectations, change, and relationships. Can you explain that?

Absolutely. That simple framework is how I structure all of my contract reviews and contract design projects.

• Managing expectations is about making sure each party knows what’s being delivered, when it’s due, what it costs, when payment is due, and who’s responsible for what. That includes scope, timing, price, liability and standards.
• Managing change means having clear processes for variations, delays, disputes, or anything unexpected. Because in the real world, things do change, and contracts need to be practical, not just theoretical.
• Managing relationships is about giving people confidence in how the contract works day-to-day, especially in grey areas. That means being clear about approvals, points of contact, who can say yes or no, and what happens if the relationship sours or trust starts to erode. We want the people involved to be able to navigate complex situations and ideally keep the relationship intact.

When those three foundations aren’t in place, things can unravel fast and small issues turn into big, costly conflicts. A good contract is one that supports the project team, not just the legal team.