The problem before the name

You don't notice you need a deal desk until the symptoms pile up. Two reps quote the same package at different prices. A discount that quietly halved your margin turns out to have been approved by no one — the rep just did it to close the quarter. A non-standard term slips into a deal and nobody realizes it commits you to something you can't deliver until delivery. Each of these is small on its own; together they tell you that pricing and terms have drifted from a decision into a habit, and habits don't protect margin.

A deal desk is simply the answer to "who decides, and by what rule, when a deal isn't standard?" In a large company it's a team. On a small team it might be one founder and a written rule. The size doesn't matter; the function does — a single, known place where non-standard pricing, discounts, and terms get reviewed before they're promised to a customer. It's the operational counterpart to the policy work in a sales discount policy that protects margin: the policy says what's allowed, the deal desk enforces it deal by deal.

Standard by default, exceptions by approval

The whole design principle is make the common case frictionless and the risky case deliberate. If every deal needs approval, reps route around the desk and it dies. If nothing needs approval, you're back to drift. The line between them is the concept of a standard deal.

Define what standard looks like: your normal packages, list pricing, standard terms, discounts within a pre-approved band. Anything inside those lines a rep can quote and close on their own, no approval, no waiting. This is why clean, agreed pricing tiers matter so much — the clearer "standard" is, the more deals never need the desk at all. Then define the triggers that do require review, and keep the list short:

  • Discount beyond the standard band. Small discretionary discounts are fine; anything past a set threshold needs a yes from whoever owns margin.
  • Non-standard terms. Custom payment schedules, unusual contract length, liability or SLA language outside your template — these are risk, not price, and deserve a look.
  • Non-standard scope or bundling. Custom configurations and one-off inclusions that your normal delivery motion isn't built for.
  • Anything that sets a precedent. A concession you'd regret giving the next customer who asks for the same thing.

Keep the approval fast or reps will route around it

A deal desk only works if it's faster than the workaround. The failure mode is a review that takes three days while the deal cools — reps learn to avoid it, and you're worse off than before. A few habits keep it quick:

  • One clear approver per trigger. The rep should never wonder who to ask. Discounts go to one person; legal terms to another. Ambiguity is delay.
  • A tight service level. Commit to a turnaround — same day for routine exceptions. Speed is what earns the desk its legitimacy, and it connects to speed-to-lead: momentum you win at the front you can lose at the quote.
  • Ask for the why, not just the what. "20% because the competitor is at that price and this logo opens a new vertical" is a decision you can make; "20% because they asked" isn't. Requiring a short rationale also improves your win-loss analysis later.
  • Record the decision on the deal. The approval, the rationale, and the terms should live on the opportunity record, not in a lost chat thread — the same activity-logging discipline that makes everything else in the CRM work, covered in logging CRM activity.

Where the CRM does the work

You don't need dedicated deal-desk software to start; you need the approval to be visible and enforced where deals already live. Because the request, the approver, and the outcome all attach to the opportunity, your pipeline stays honest about what was actually promised — and workflow automations can route an exception to the right approver and flag deals that skipped the step, so the process holds without a human policing it.

The payoff compounds quietly. Margins stop leaking through unapproved discounts. Quotes become consistent enough that customers can't play reps against each other. And when you sit down to do price-to-win or renewal planning, you're working from real, deliberate pricing history instead of a scatter of one-off concessions nobody decided on. A deal desk isn't bureaucracy — it's the difference between pricing that's chosen and pricing that just happened.