How would I measure internal delay if I wanted to start today?

Pick one closed deal and reconstruct it from the email thread. For each item, write down the date it arrived and the date someone accepted it. The gap between those two dates is internal delay, and it is usually the first time anyone at the firm has seen it written down. It is tedious for one deal and impossible for twenty, which is the actual reason nobody measures it.

Isn't chasing our own colleagues going to be unpopular?

It is less unpopular than it sounds, because the alternative is a partner asking in person, which is worse for everyone. A queue that says an item has been with Legal for four days is not an accusation — Legal usually knows, and often the item is stuck behind something nobody told them about. The information is what is missing, not the willingness.

What is a reasonable target for open to IC-ready?

There is no credible published benchmark, and any firm quoting one at you should be asked where it came from. The number that matters is your own baseline measured the same way twice — first before you change anything, then after. A target invented before you have measured yourself is a number you will quietly stop reporting.

The half of diligence delay that happens inside your own building

Ask a deal team why the last deal took eleven weeks and you will hear about the borrower. They were slow, their finance director was on holiday, their accountant sat on the FY24s for a fortnight.

All of that is usually true, and it is at most half the story. The other half happened inside your own building, and nobody wrote it down.

The gap nobody looks at

There are two clocks running on every requirement.

The first starts when you ask for something and stops when it arrives. That one is visible: it is the one people complain about, the one that generates follow-up emails, the one a chasing process is aimed at.

The second starts when the document lands and stops when somebody accepts it. It is invisible, because nothing is watching it. A file arrives, sits in a folder or a thread, and at some point a reviewer gets to it. There is no artefact anywhere in the process that records how long that took.

The result is a process that is instrumented on the outside and blind on the inside — and it is the inside half you have any control over.

Why it stays invisible

Three reasons, and none of them are laziness.

The work has no queue. "Legal to review the facility agreement" is a real piece of work that must be done before the file is complete, and on most deals it exists only as an expectation in someone's head. It is not on the checklist, because the checklist is a list of things the counterparty owes. So there is nothing to be late.

The state has no name. When a document arrives, most processes have one state for it: received. But "a file arrived" and "we think this file answers item 14" and "a reviewer has accepted it" are three different facts, and a process that collapses them cannot tell you which of the three a given item is sitting in.

Asking is socially expensive. Chasing a borrower is a normal professional act. Chasing a partner is not, quite. So internal items get chased in person, at the wrong moment, by whoever is most anxious about the deal — which is a worse experience for everyone than a queue would be.

What it looks like when you do measure it

The pattern most firms find, once they can see it, is not one enormous stall. It is a scatter of three-and-four-day gaps that nobody would have called a problem individually, spread across twenty items, with one genuine week-long block sitting behind something a reviewer was waiting on and had not mentioned.

That distribution matters, because it changes the fix. If the delay were one big stall you would solve it with escalation. Being a scatter, it is solved by visibility: the four-day gaps mostly close on their own once someone can see them, and the one real block turns out to be a dependency nobody had named.

The other common finding is that the bottleneck is not where the deal team assumed. Legal gets blamed for a great deal of delay that is actually a document going back and forth because the first version was unsigned.

Making it visible without making it adversarial

A few things that help, in rough order of how much they help.

Put internal work on the same list. If Legal's review is a checklist item with a state and an owner, it is late in the same way and for the same reasons as anything else. It stops being a favour and starts being work with a position in a queue.

Separate "arrived" from "accepted". Two states, not one. Then "we have nine items sitting with us" is a sentence anyone can say, and it is nobody's fault in particular, which is exactly what makes it sayable.

Chase internally on the same cadence. Not harder than the counterparty — the same. The asymmetry is the tell: a firm that nudges a borrower every four days and its own Legal team never is telling you where it thinks the problem is, and it is wrong half the time.

Show what happens next before it happens. If a partner can see that after Credit signs it goes to Legal and then to IC, they stop asking, and the person who would have answered stops being interrupted. Most status questions are not requests for information; they are requests for reassurance that the thing has not been dropped.

The number to hold on to

Days from opening a deal to IC-ready.

It is the only measure that includes both clocks. Time-to-response measures the counterparty. Time-in-review measures you. Only the total measures the process, and it is the one an investment committee actually feels.

Measure it before changing anything, so the baseline is real. Then measure it the same way afterwards. Any firm quoting you an industry benchmark for it should be asked, politely, where the number came from — there is no credible published figure for pre-close diligence cycle time, and a benchmark invented to win a meeting is worth exactly what it cost to produce.