The approval that took four days
Most price losses are clock losses. The approval took four days; the buyer's quarter closed on day three. The fix is tiered authority, written down.
9:14 on a Thursday night. A message lands in a channel with six people on it: “Buyer needs 18% to get this through their board Monday. Can we do that?”
Four thumbs-up reacts. No answer.
At 7:40 the next morning, someone replies: “I think that’s above my line?” Question mark included. By noon the thread has gained two people and lost one to PTO. By the following Tuesday there is a decision, and the decision is yes — the same yes that was sitting there on Thursday night.
The buyer’s board met on Monday.
I’ve watched that thread play out at more than one company. The details change. The shape never does. Somebody asks a reasonable question, nobody is certain it’s their call, and the deal waits while the org works out who gets to decide.
“We lost on price” is usually a clock problem
Illustrative — not a client, not a real deal, just the math
Rep asks for an exception on a competitive displacement. The buyer’s fiscal quarter closes Wednesday.
First approver is traveling. Sees it, doesn’t action it.
Second approver wants the margin impact before signing off. That number lives with someone in finance who is in back-to-backs until Wednesday.
The buyer’s committee meets. There’s no number in front of them. They go with the vendor who answered.
Nothing in that sequence is about the discount. The number was almost certainly approvable on Friday afternoon — the company just couldn’t say so out loud in time. The loss report says price. The loss was availability.
This is the part a win/loss review miscodes. A discount request isn’t a pricing question. It’s a question with a deadline attached to somebody else’s calendar, and that calendar is not negotiable. Your buyer’s quarter closes whether or not your approval chain is ready. Their board meets whether or not your second approver is out of back-to-backs.
Four days is not a long time in most parts of a business. In the last week of somebody else’s quarter, four days is the whole game.
And those four days never show up anywhere you’d think to look for them. There’s no field in the CRM called waiting on us. The stage timestamps show the deal sitting in negotiation, which reads like a buyer deliberating. Sometimes it is. Often it’s your own approval chain, and the report can’t tell the difference.
The record nobody wants until they badly need it
The second failure is quieter, and it doesn’t bill you on the day it happens.
The approval eventually landed. It landed in a direct message. Or in a hallway, confirmed later by a rep who wrote “approved by [name]” in the CRM notes and moved on. Or in a channel that got archived in a workspace cleanup. There’s no threshold it was measured against, no reason attached, and no expiry.
Three months later, the questions start.
The renewal comes up at the discounted rate, and nobody can tell whether that was a one-time concession to close a quarter or a new floor the customer is now entitled to. A second buyer in the same segment asks for the same number, and you can’t articulate why the first one earned it — so you either give it away again or defend a line you can’t source. Finance asks why average selling price keeps drifting, and the honest answer is that a dozen separate people each made a defensible call and nobody was watching the aggregate. The rep who asked for it has moved teams. The manager who approved it remembers approving something.
An approval without a recorded reason isn’t an approval. It’s a memory.
Memories don’t survive a reorg.
And in a regulated or audited sale — clinical research, health systems, insurance, financial services — “we’re fairly sure someone approved it in a DM” is not a sentence you want to say out loud. I built a Deal Desk from scratch governing a $25M+ pipeline, with the approval logic wired through the tools we already owned — Salesforce CPQ and DocuSign. The rules weren’t the hard part. The hard part was the one-line reason field, because that’s the field that makes the record worth keeping.
Competitive displacement, incumbent renewal in 60 days, three-year term.
Nine words. Three months of arguments avoided. Compare: ~~”To win the deal”~~ — not attribution. A shrug with a timestamp.
A good reason traces to something specific — a concern the buyer raised, a constraint you were working around, a piece of feedback, an issue in the deal. That is what makes it useful later: you can read it back, see what actually drove the concession, and refine the approach next time. A generic reason gives you nothing, and a year of those teaches you nothing about your own pricing.
What tiered authority actually is
Here’s the boring truth: most companies already own the tool. The approval routing in your CRM or your quoting system is sitting there unconfigured, because configuring it requires a decision nobody has made — who is allowed to approve what.
Tiered approval authority is three questions answered in advance and written down.
Who can approve what. Not “who is senior.” Who owns the exposure. A rep can commit the routine concession inside their own authority without asking anyone. A manager owns the next band. The deal desk owns the exceptions.
At what threshold. Real numbers, published, not vibes. In enterprise software the bands I have worked with run wider than people expect, and the ownership map matters more than the percentages:
| Band | Who owns it |
|---|---|
| Up to ~10% | The rep, inside their own authority, logged |
| 10–25% | Regional or sales director |
| 25–50% | Commercial leadership — senior director, VP, or CRO |
| 50%+ | Executive — CEO, CRO, CFO territory |
Two things almost everyone skips here. A margin floor — a number below which nobody approves alone, ever, no matter their title. And an expiry — an approval granted in March is not still live in July. Approvals that don’t expire become precedent, and precedent becomes list price.
With what recorded reason. One sentence, captured where the approval happens, not in someone’s sent folder. That sentence is the entire audit trail. It’s also the input to every discount analysis you’ll ever want to run.
Then the part that actually kills the four-day problem: a named backup at every tier. Authority that lives in exactly one person isn’t authority. It’s a bottleneck with a title.
What this is not is another layer. Done properly, tiered authority removes approvals. Most requests stop being requests at all — they become decisions someone was already authorized to make, logged on the way past. The volume hitting a senior inbox drops hard, which is the only reason the genuine exceptions ever get a real look.
The fix is boring, and that’s the point
Nobody gets promoted for writing an approval matrix. It’s a page and a half. It takes an afternoon and two arguments to write, and about three weeks of enforcement before anyone trusts it. The genuinely hard part comes in week four, when a large deal asks for an exception to the exception process and someone senior is inclined to grant it. Hold the line there and the thresholds are real. Fold once and you’re back to direct messages.
The principle underneath it is the one I keep coming back to: oversight proportionate to exposure. Not every answer, approval or bid carries the same risk, so they shouldn’t all carry the same review.
Most approval processes get this exactly backwards. They apply uniform review to wildly different exposure. A 4% discount on a small renewal takes the same trip through the same three inboxes as a 30% multi-year displacement against an incumbent. That’s precisely why the routine ones take four days — the queue is clogged with decisions that never needed to be in it, so the one decision that genuinely warranted senior attention is stuck behind forty that didn’t.
Review everything and you review nothing well.
So the routine gets to move at the speed of the person closest to it, with a record. The genuine exception escalates, with a record. Same system, two speeds, sorted by risk rather than by habit.
And here’s the reframe worth keeping: the point of governance isn’t control. Most sales leaders hear “approval governance” and picture a longer form and a slower deal. It’s the opposite. The point is that nobody has to wait for one specific person to be online. Speed is what governance buys you. Control is just what it looks like from the outside.
The four-day approval isn’t a discipline problem, and it isn’t a tooling problem. It’s an unmade decision, and it gets remade under deadline pressure on every single deal — which is the most expensive possible time to make it.
Make it once, on a quiet Tuesday, with nobody’s quarter on the line.
Pull your largest discount from last quarter. Can you reconstruct, from a record and not from someone’s memory, who approved it and what reason they gave?
Rob Roberts is the founder of Widenine Solutions. In B2B SaaS and life sciences since 2011, running Deal Desk and Sales Operations since 2018 — including a Deal Desk built from scratch governing a $25M+ pipeline, and a 2,000+ vetted Q&A library behind 5,000+ responses built and audited.