How to Measure Contract Cycle Time Without a CLM

Ask a legal team how long their contracts take and you'll usually get one of three answers: a shrug, an anecdote about the worst deal of the quarter, or a number somebody made up in a QBR two years ago and nobody has questioned since.

That's not a criticism of the team. It's a symptom of how contracting works. The request arrives by email, the negotiation happens in tracked changes, the approval happens on Slack, and the signature happens somewhere else entirely. No single system saw the whole thing, so no single system can report on it.

The good news is that you don't need to buy anything to fix this. You need about an afternoon and a willingness to look at the answer honestly.

Start with the last forty deals, not the last four hundred

The instinct is to measure everything. Resist it. Forty to sixty recently closed agreements is enough to see the shape of the problem, and it's small enough that you'll actually finish.

Pick deals that closed in the last two quarters. Closed deals have a definite end date, which open ones don't, and recent ones reflect how you work now rather than how you worked before the last reorg.

The two timestamps you need

For each deal, find two dates:

The request date. When did the business first ask legal for the contract? Not when legal opened it. Not when it landed in the queue. When the ask was made. This is almost always in email or a ticketing system, and it is almost always earlier than legal thinks.

The signature date. When was the last signature applied. This one is usually easy, since it's in your e-signature tool or the CRM.

The gap between them is your cycle time. That's the whole measurement. Everything else is refinement.

Report the median, not the average

One catastrophic deal that took 94 days will drag your average into fiction. The median tells you what a typical deal actually looks like, which is the number you can manage against.

Report the median by contract type, too. A mutual NDA and a master services agreement are different animals, and blending them produces a number that describes neither. If your NDAs take eleven days, that is a finding. Burying it inside an "all contracts" average hides it.

The split that makes it actionable

Here is the part most teams skip, and it's the part that matters.

Take your cycle time and cut it in two:

Request to first redline. This is legal's lane. If this segment is long, you have a queue and capacity problem: too much volume, not enough reviewer time, no prioritization rules.

First redline to signature. This is almost never legal. This is approval authority, procurement, security review, finance, and the question of who is actually allowed to say yes. If this segment is long, hiring another lawyer will not help you.

I have watched teams spend a year building a business case for additional legal headcount when 70% of their cycle time was sitting in the second bucket. The headcount would have been approved, spent, and made no measurable difference to the number the CRO cared about.

Measure the split before you propose the fix.

Three more numbers worth having

Once you have the baseline, these sharpen it:

Negotiation rounds per deal. How many times did paper go back and forth? Two rounds is healthy. Five means your first draft is starting a fight it doesn't need to start.

The clauses causing those rounds. Read the redlines on ten deals and tally what got marked up. It's usually the same three clauses over and over: liability cap, indemnity, and termination. That tally is a map of exactly what your playbook needs to pre-approve.

Percentage that never touched legal. If the answer is zero, every routine NDA in your company is queuing behind work that actually needs judgment. That's a policy problem, not a volume problem.

What to do with the number once you have it

Two things.

First, reconcile it to what sales already tracks. Your CRM has stage-duration data. When your cycle-time number and the sales team's "stuck in legal" number describe the same reality, you stop arguing about whether the problem exists and start talking about what to do. That shared number is the entire basis for getting the fix funded out of a revenue budget rather than a shrinking legal one.

Second, write it down and date it. You are going to change something. When you want to prove the change worked, you will need the before. Teams that skip this step end up making real improvements they can never demonstrate, which is how good work gets defunded.

The honest caveat

This measurement is manual, and manual measurement decays. You will do it beautifully once and then not again for eight months.

That's fine. The point of the baseline is not to build a permanent reporting practice. It's to find out where the time actually goes, so the next thing you do is aimed at the right target. A one-time honest number beats a permanent dashboard measuring the wrong thing.

---

If you want a faster read on where your process stands before you pull any data, the Deal Velocity Scorecard is twelve statements covering baseline, authority, and friction. It takes about four minutes and tells you which of the three is your actual bottleneck.