Who approves discounts in food and beverage, and why approval rates are not a measure of control
In most food and beverage businesses, discount approval runs upwards. A rep requests, a sales manager approves, and above a set value the request goes to a commercial director and then to finance. On paper each step sits further from the customer and closer to the P&L. In practice each step is measured on the same revenue number the deal helps deliver, which makes approval a commercial decision taken by an interested party rather than an independent check. That is why approval rates in most businesses run very high, why approvals cluster in the closing week of a period, and why a healthy compliance summary can sit alongside disappointing account margin. Independence, not seniority, is what makes an approval route a control.
Who approves discounts in a food and beverage business?
In most cases, the person one level above the person requesting. A rep requests, a sales manager approves. Above a set value the request escalates to a commercial or sales director. Above that, finance.
The design assumes that each step upwards adds distance from the customer relationship and therefore adds objectivity.
The assumption holds for distance. It does not hold for interest.
The rep has a target. The sales manager carries the sum of the team’s targets. The director carries the region. The deal waiting for approval is part of the number that every approver in the chain is judged on.
That does not make anyone dishonest. It means the deal is reviewed by people who need it to close, at the point in the period when they need it most.
Why is a discount approval route rarely independent?
Because seniority and independence are not the same thing.
An independent approver is someone whose own measurement does not improve when the deal goes through. In a conventional sales hierarchy, nobody in the route meets that description until the request reaches finance, and by then most requests have already been resolved lower down.
The practical test is simple. Ask what the approver’s bonus is calculated on, and whether the account in question sits inside it. If it does, the approval is a commercial judgement about whether the concession is worth making, not a check on whether it should be made at all.
Both are legitimate activities. They are not the same activity, and only one of them is a control.
What does a high approval rate actually tell you?
Take a quarter with several hundred discount requests and fewer than ten declines. Presented as a compliance summary, that reads as a well-run process with strong policy adherence.
Read the other way, it means almost nothing that entered the route met any resistance.
A control that approves nearly everything is a record rather than a control. It captures what was agreed. It does not shape what gets agreed.
The more useful figure is the decline rate broken down by approver, because that shows whether the route is applying judgement consistently or simply processing requests. A team where one manager declines occasionally and the rest decline never has a governance question, not a data question.
What is corridor approval, and why does it not show in the audit trail?
A rep catches their manager on the way out and asks whether a particular level will fly. The answer is yes. The formal request goes into the system afterwards, already agreed.
The audit trail shows a request and an approval within the hour, which reads as a responsive process.
What it does not show is that the decision was made before the request existed, and that the conversation was about a monthly number rather than about the account.
This matters for reporting more than it matters for the individual deal. Any analysis of approval behaviour is drawn from timestamps, and timestamps record when the system was updated, not when the decision was taken.
Why do managers restructure a deal rather than escalate it?
Escalation carries a cost that restructuring does not.
Taking a request upwards means telling a director that a customer is under pressure and the number may not land. That conversation has consequences beyond the deal.
Reshaping the deal so it sits just underneath the authority limit costs nothing and keeps the conversation local. Most managers restructure.
The effect is a distribution of agreements bunched immediately below each authority level, and a senior team who see very few of the deals the escalation route was built to surface.
Why do approvals cluster at the end of a period?
Because that is when the pressure on the approver is highest and the time available to interrogate the deal is lowest.
Approvals are rarely spread evenly across a month or a quarter. They concentrate in the closing week, when the approver has the strongest personal reason to say yes and the least capacity to ask a second question.
Comparing the volume of approvals in the opening week of a period against the closing week is one of the quickest tests available, and it uses data that already exists.
Why do two managers give different answers to the same request?
Same discount, same category, two different outcomes depending on who signs.
Reps learn this quickly. Customers learn it too, usually through their own network rather than through anything anyone says.
Over time, pricing consistency across the customer base starts to depend on who happened to be available, which creates a second problem. Two similar accounts on materially different terms is difficult to defend when the customers compare notes, and difficult to unwind once both have settled into their position.
What do directors and finance actually see?
Directors rarely review deals. They review summaries: approval volumes, average discount, compliance against policy.
Every one of those measures is drawn from the deals that entered the approval route in the first place. That is why the summary can look healthy while account margin does not.
Finance sees the real position, but sees it last. Credit notes, promotional contributions, extended payment terms and free-of-charge stock arrive weeks after the agreement, through different systems, recorded against different reasons. By the time they can be assembled into a view of what a customer actually costs, the quarter is closed and the agreement is already running.
None of this is a compliance failure. Nobody is turning a blind eye. Each layer is doing what it is measured to do.
What happens to a deal that gets declined?
This is the question most approval reporting does not answer.
A decline is only a control if the concession stops. If the customer still receives an equivalent outcome through a route the approval process does not cover, the decline has moved the cost rather than prevented it.
Tracking what an account received in the period after a declined request is the single most revealing check available, and almost nobody runs it.
How do you test whether your approval route is a control?
A short exercise, using approvals already recorded rather than a new report.
Take last quarter’s approvals and sort them two ways. First by approver, comparing decline rates across the team. Then by date, counting how many landed in the final week of each month against the opening week.
After that, take the five accounts with the worst margin variance and check who approved their pricing, and what that person’s target was.
Four questions are worth putting to your own approval route.
- What proportion of requests were declined last quarter, broken down by approver.
- Whether each approver carries a revenue target that includes the account in question.
- How many approvals landed in the closing week of the period, compared with the opening week.
- What happens to a deal that is declined, and whether anyone tracks what that customer received instead.
If the honest answer to the last one is that nobody checks, the decline is not a control either. It moves the concession somewhere the approval route cannot see.
What does a stronger approval route look like?
Three things tend to matter more than the authority levels themselves.
The first is genuine independence somewhere in the route. At least one approver whose measurement does not improve when the deal closes, involved early enough to influence the outcome rather than to observe it.
The second is coverage. Every concession that carries a cost belongs in the same framework, whether it is expressed as a percentage, a payment term, free stock or a promotional contribution. An approval route that governs only percentage discounts governs only the concessions people choose to express as percentages.
The third is a single account view. Everything a customer received in a period needs to roll up to one record, including sites trading under different codes within the same group, so the approval summary and the margin position can be read side by side rather than months apart.
How Allsop helps food and beverage businesses close the gap
At Allsop, we work with food and beverage businesses through intelligent software for Customer Order Management, Data Workbench and Customer Margin Management.
Customer Order Management routes orders and amendments through a controlled process, so restructured deals, split orders and after-the-fact adjustments are visible rather than absorbed.
Data Workbench improves the process behind master data, so customer hierarchies, product records and agreed price records stay accurate and consistent. That is what allows every concession an account receives to roll up to the right customer, including group structures trading across multiple sites and codes.
Customer Margin Management brings the agreed terms, the invoiced position and the later adjustments together, so account margin can be read as one figure rather than reconstructed from four systems after the quarter has closed.
We do not decide what a customer should pay, and we do not set approval policy. We work on the process around those decisions, so approval routes reflect who is genuinely independent of the deal, exceptions are recorded where they can be reviewed, and the full cost of a customer agreement sits in one place.
If your approvals look disciplined and your account margin does not, the distance between those two things is worth understanding. Speak to us, or read more about how we approach margin protection across the order to invoice process.
Frequently asked questions
Who should approve discounts in a food and beverage business?
Approval works as a control when at least one person in the route is independent of the outcome, meaning their own measurement does not improve when the deal closes. In a conventional sales hierarchy every layer carries a version of the same revenue target, so seniority alone does not create independence. Involving a function measured on margin rather than volume, early enough to influence the deal, gives the route more weight.
Is a high discount approval rate a good sign?
Not on its own. A route that approves almost everything submitted to it is recording decisions rather than testing them. The more informative measures are the decline rate broken down by approver, and the distribution of approvals across the period, since approvals concentrated in the closing week indicate pressure rather than process.
Why do discount approvals cluster at the end of the month or quarter?
Because that is when the approver has the strongest commercial reason to agree and the least time to question the detail. The approver and the requester are usually measured on the same period-end number, so the point of greatest pressure on the deal is also the point of least scrutiny.
What is corridor approval?
An informal agreement reached in conversation before the request is entered into the system. The formal record then shows a request and an approval within a short window, which reads as an efficient process. The audit trail captures when the system was updated rather than when the decision was made, so this behaviour does not appear in approval reporting.
Why do two managers approve the same discount differently?
Because authority levels set a limit but rarely set a standard for what qualifies. Without shared criteria, outcomes vary by individual judgement and by the pressure that individual is under. Over time this produces similar customers on materially different terms, which becomes difficult to defend and difficult to unwind.
How do you check whether a discount approval process is working?
Sort last quarter’s approvals by approver to compare decline rates, then by date to compare the closing week of each period against the opening week. Take the accounts with the worst margin variance and check who approved their pricing and what that person was measured on. Then check what declined customers received in the following period, since a decline that moves the concession elsewhere is not a control.
