How do you build discount governance that actually sticks — what combination of policy, tooling, and incentive alignment prevents reps from circumventing rules through bundling tricks in 2027?

What discount governance actually is, and why the bundling loophole is the real test

Most companies think they have discount governance because they have a discount policy. They do not. A policy is one input to a governance system; governance is the whole equilibrium — the rules, the enforcement, the incentives, and the maintenance cadence that keeps all three from drifting apart. The distinction matters most at the exact point where governance is usually tested and usually fails: the moment a rep who has hit a discount ceiling starts restructuring the deal rather than accepting the ceiling.

That restructuring is what people mean by "bundling tricks," and it is worth being precise about the mechanics, because vague talk about reps "gaming the system" produces vague controls. The specific moves recur across every B2B seller I have seen:

Line-item shell games. The policy caps discount at the line level, so the rep discounts one SKU to near zero and raises another to keep the blended number inside the band. The CPQ price rules, which evaluate per-line, see nothing wrong. The blended effective discount is 12 points deeper than policy allows.

Free-attach. The rep holds list price on the licensed product — the number governance actually measures — and gives away onboarding, premium support, a professional services block, or an extra module "at no charge." Zero recorded discount. Real margin transfer, sitting in a services cost center nobody is measuring against the deal.

Term-and-volume stacking. The policy grants extra discount for a three-year commitment and extra discount for a volume tier. The rep takes both, then adds a ramp that makes year one tiny, then negotiates a mid-term out clause. What the policy priced as "three years of committed volume" is functionally a one-year deal at a three-year discount.

Deal splitting. A $400K deal that would route to the CRO becomes two $200K deals that each clear at the VP tier, signed a week apart, same legal entity.

The side letter. The quote in CPQ is perfectly compliant. The commercial reality lives in a countersigned amendment, a redline in the MSA, or an email from the rep promising a credit next renewal.

Uplift suppression. The rep holds the current-year price but negotiates away the contractual renewal uplift, moving the concession into a future period where nobody's discount dashboard will ever see it.

Every one of these is a rational response to a system that measures one number — headline discount percentage on the licensed line — while paying the rep on a different number: bookings. The rep is not being dishonest. They are optimizing exactly what you built. If you take one thing from this page, take that: circumventing behavior is a design output, not a character problem. When a governance system produces widespread creative structuring, the correct diagnosis is that the measured surface is narrower than the concession surface, and the incentive points away from the policy.

This is why the answer to the question is a *combination* rather than a single control. Policy alone defines a surface reps can route around. Tooling alone defines a checkpoint reps can restructure past. Incentive alignment alone, with no specific rules and no enforcement, is a sentiment that evaporates the first time someone is 80% to quota in week twelve. The combination works because each leg closes the hole the others leave open: the policy names bundling as a violation in its own right, the tooling measures blended and total-contract economics instead of per-line headline discount, and the comp plan makes the free services block cost the rep personally.

There is a broader RevOps pattern hiding here worth naming, because it recurs far outside pricing. Any control that measures a proxy rather than the outcome creates an arbitrage between the proxy and the outcome, and a compensated actor will find the arbitrage faster than you can close it. Forecast-category rules get gamed by sandbagging. Activity metrics get gamed by low-value logged calls. MQL thresholds get gamed by loosening the scoring model. Discount governance is simply the version of this problem where the arbitrage is denominated in gross margin, which is why it hurts the most.

Building the policy layer so the rules survive contact with a live deal

The bar for a real discount policy is inter-rater reliability: two deal desk analysts, reading it independently, reach the same decision on the same deal. If your document cannot pass that test, it is a sentiment, not a policy. Six components get you there.

The margin floor, owned by finance. This is different in kind from everything else. Bands and thresholds are guardrails — crossable with the right approval. The floor is a floor. It exists because at quarter-end the interests of the sales organization and the interests of the business genuinely diverge, and something has to represent the business at the moment that interest is under maximum pressure. If sales owns the floor, it is not a floor; it is another guardrail that bends. Express it as an effective-margin number, not a discount percentage, precisely so bundling cannot get underneath it: a deal at 20% headline discount with $60K of free services can be below floor while a deal at 35% headline discount with no attach sits comfortably above it.

A two- or three-dimensional authority matrix. "Discounts over 20% need VP approval" is a slide. A real matrix crosses discount depth with deal size, and often with segment or deal type. A 15-unit deal at 18% clears the fast lane. The same 15-unit deal at 28% routes to a regional VP with a mandatory justification field. A 150-unit deal at that identical 28% routes to the CRO and deal desk jointly, because the dollar exposure is an order of magnitude larger even though the percentage is the same. Percentage-only matrices are the most common structural defect in discount policy, and they are the reason large deals get under-scrutinized relative to their risk.

Standard bands. Pre-approved discount levels tied to deal size and term, so the ordinary case is explicitly blessed and needs no approval at all. Without standard bands, every discount is an exception, the deal desk drowns, and the fast lane is impossible to build. Standard bands are what make governance tolerable to the people living inside it.

Multi-year and volume rules with explicit anti-stacking language. Say what a two-year commitment earns and what a three-year earns. Then say, in the same paragraph, what stacking does *not* earn: that term and volume concessions are not additive beyond a stated cap, that a ramp schedule with a below-threshold year one does not qualify for the multi-year band, and that any termination-for-convenience right voids the multi-year discount entirely. This one paragraph closes the single most common bundling route.

Total-concession language. Define discount as the full economic concession, not the licensed-line percentage. Name the components: license discount, waived or discounted services, free modules or seats, extended payment terms, suppressed renewal uplift, and any credit or make-good. State that all of them roll into one governed number. This is the policy sentence that makes free-attach a violation rather than a clever workaround, and its absence is why most policies fail the bundling test on paper before they ever fail it in practice.

Named exception criteria and never-negotiables. Write down the narrow set of circumstances that justify going below band — a reference logo in a new segment, a competitive displacement with clear expansion math, a partnership with named strategic value. The purpose is not to enable exceptions; it is to *contain* them, so "strategic" becomes a defined category rather than a word meaning "I want to discount more." Then list the short set of things not on the table for anyone: below-floor without finance, concessions not tied to a corresponding customer commitment, retroactive credits, and anything undocumented in the system.

The specificity test is mechanical and it is the bridge to the next leg: can your CPQ admin build it? If they come back and say the rules are ambiguous, the policy is not finished. The act of trying to encode it forces the precision that makes the rule mean the same thing to everyone, every time — which is the precondition for holding anyone to it fairly.

The step-by-step process for standing it up

Sequence matters more than people expect, because the legs are load-bearing for each other. Encode before authoring and you encode ambiguity. Launch before fixing comp and you launch a system the comp plan actively fights, and reps learn to game it before you get around to the fix. Launch before enabling managers and the message dies in the first one-on-one.

Walking the sequence in more detail:

Baseline before you build. Pull four quarters of closed-won deals and compute realized price against list, including services and credits. Most teams discover the real number is 6–15 points worse than the discount field says, and that gap *is* the bundling leak quantified. It is also the business case for everything that follows, and you will need it to justify the CPQ spend and the deal desk headcount.

Author, then test for encodability. Finance sets the floor. RevOps and deal desk build the matrix, bands, anti-stacking rules, and exception criteria. Then a CPQ admin reads it with one question: can I build this? The loop back is normal and healthy — expect two or three passes.

Encode with the fast lane as a first-class requirement, not a phase two. Price rules translate bands and floor into constraints. Hard blocks make the floor genuinely hard: not a warning a rep clicks through, a stop. Approval routing turns the matrix into workflow with justification fields enforced before submission. The audit trail captures depth, type, justification, and term linkage as structured queryable fields rather than free text.

Fix comp before launch. This is the step companies defer and the deferral is fatal.

Enable managers, then launch behind visible leadership modeling. The launch is not an email. It is leadership putting their own deals through the process, repeatedly, including when it costs them.

Measure from day one and calendar the drift review before you need it.

A reasonable timeline: policy authoring two to four weeks of concentrated work; CPQ encoding four to twelve weeks depending on whether you already have a CPQ, with a greenfield implementation running a quarter or more; comp realignment aligned to a plan-year boundary, which is often the real constraint on the whole program; manager enablement two to three weeks running in parallel. Most teams should plan a full quarter minimum and target a plan-year start for launch.

Where the tooling layer earns its keep, and where teams under-invest

The principle is absolute: the policy lives in the tooling or it does not live. A policy that exists only as a document depends on every rep, every time, under every pressure, choosing to comply. That is not governance. That is hope with a version number.

What tooling changes is the *default*. In a policy-only world, the default is whatever the rep chooses and compliance is the effortful exception. In a properly tooled world, the default *is* the policy — correct routing, hard floor, captured data, automatically — and circumventing it becomes the effortful path. Governance is largely a question of which direction the friction points.

Which brings up the most violated principle in the whole discipline: governance that is all friction gets routed around. If every quote, including the perfectly ordinary in-policy quote, waits in an approval queue, the sales organization experiences governance as an enemy of velocity, and they are correct. A system that punishes the compliant rep as heavily as the non-compliant one has destroyed its own legitimacy.

The fast lane is the fix. An in-policy quote — standard band, normal size, normal term — should generate and be sendable with zero human approval steps, because the policy already pre-blessed that shape of deal. Friction is *reserved* for exceptions. This does two underrated things. It makes staying in policy the fastest path to a sendable quote, so the rep's convenience and the company's margin point the same way. And it concentrates the deal desk's scarce attention on deals that carry actual risk — a desk drowning in routine approvals cannot scrutinize the genuinely dangerous ones. Speed for the compliant, scrutiny for the exceptional. A governance system without a fast lane has a half-life measured in quarters.

For the bundling problem specifically, the tooling has to measure the right surface, and this is where most CPQ configurations fall short:

Evaluate blended and total-contract discount, not just line-level. Configure a quote-level calculated field that computes effective discount across every line including services, then route on *that* number. Line-level-only evaluation is the direct enabler of the shell game.

Give services a real internal cost. If professional services and onboarding carry a standard cost in the model, a "free" implementation shows up as margin erosion in the deal's calculated economics rather than as zero recorded discount. This one configuration change closes free-attach more effectively than any amount of policy language.

Model the ramp on total contract value, not year one. Evaluate a ramped deal against its full-term committed value so a tiny year one cannot buy a multi-year band.

Flag split-deal patterns. An alert on multiple opportunities against the same account or parent entity closing within a rolling window catches threshold splitting. It does not need to block — a flag to the deal desk is enough, because the behavior mostly stops once reps know it is visible.

Require the paper to match the quote. Route non-standard contract language through the same approval path as the discount, and make CLM redlines visible to the deal desk. A compliant quote with a non-compliant amendment is the loophole that no pricing rule can see.

The predictable under-investments: no CPQ at all, where quotes live in spreadsheets and governance is honor-system; approval-by-email, which feels like governance while providing no fast lane, no hard block, and an audit trail scattered across inboxes; and the spreadsheet price book that forks, drifts, and leaves reps discounting off inconsistent baselines. All three come from the same false economy — the policy document is nearly free to write, the tooling is not, so companies write the document, declare victory, and discover eighteen months later it never had teeth.

One note on the audit trail's framing, because it determines whether the data is real. Position it as "this is how we learn and tune the policy," not "this is how we catch you." If the org experiences it as surveillance, the real commercial action moves off-system into side letters and hallway commitments, and your trail faithfully records a fiction. Use it in the quarterly review to adjust bands, not in one-off witch hunts. When a rep's deal surfaces as an exception, the opening line is "help me understand the deal," not "you broke a rule." Visibility is enforcement, but it works through transparency, not fear.

Costs, timelines, and what the numbers usually look like

Real budgeting, with the caveat that ranges vary enormously by company size, CPQ platform, and existing data hygiene — treat these as shape rather than quotes.

CPQ. Per-user subscription pricing is standard, and total cost is dominated by implementation rather than license. A mid-market implementation on top of a clean CRM with a stable price book is a multi-week project; a complex catalog with usage-based components, multi-currency, and channel pricing runs a quarter or more and typically needs a partner. The budgeting error is assuming license cost is the cost. Implementation, price-book cleanup, and the integration work to make services costs visible usually exceed it.

Deal desk headcount. Costs scale with exception volume, which is a direct function of how well-calibrated your standard bands are. Bands set too tight generate exception volume that requires headcount you did not budget for; this is a common and expensive mistake, and it also poisons the culture, because reps correctly perceive that ordinary deals are being treated as exceptions. If your fast-lane percentage is low, look at band calibration before hiring.

RevOps time. Policy authoring is a few weeks of concentrated senior effort. The quarterly drift review runs a couple of days per quarter of analysis plus a working session. Ongoing tooling maintenance — new SKUs, price changes, matrix updates — is a recurring slice of someone's role, not a one-time cost, and pretending otherwise is how tooling decay starts.

Comp redesign. The cost here is mostly political and calendar-bound rather than financial. Realigning to a margin-based or margin-modified plan means a plan-year boundary, sales leadership buy-in, modeling of rep earnings under the new plan against historical deals, and clear communication. Model it carefully: a plan that materially cuts earnings for reps who were behaving reasonably under the old rules will cost you people, and the attrition will be blamed on governance.

Timeline to see the number move. Do not expect the realized-price line to improve in the first quarter after launch. Pipeline in flight was quoted under the old rules, and the honest measurement window is two to three quarters. The exception is quarter-end concentration, which responds fast because it is behavioral — if leadership holds the line at the first quarter-end after launch, you will see it in that quarter's data, and if they fold, you will see that too.

Typical distribution shapes. A healthy discount distribution is concentrated within the standard band with a thin, controlled tail. The pathology to look for is bimodality: a pile of deals near list and a pile at deep discount, which averages to a perfectly respectable-looking number while hiding a broken system. Averages lie about discounting more reliably than about almost any other sales metric, which is why every metric below is distributional or trended rather than a single mean.

The six stickiness metrics worth standing up from day one:

Discount distribution health — the shape, not the average. List-to-effective price ratio trend — realized against list over time, including services and credits, which is the number bundling actually moves. Exception volume — count and percentage through the exception path; if it is a third of deals, "exception" has become "process." Quarter-end concentration — the share of the quarter's discounting in the final ten days, the single most diagnostic number because it measures the system under maximum load. Approval cycle time, split fast-lane versus exception — the fast lane should be near-instant or reps will route around it. Fast-lane percentage — the share clearing without human approval, which tells you whether bands are calibrated to reality.

Add a seventh if bundling is your specific problem: services attach margin, tracked per deal. A rising volume of services delivered at or below cost alongside flat headline discount is free-attach showing up in the only place it is visible.

Where teams get it wrong

They build one leg and expect a stool. The most common failure by a wide margin. The policy company writes an excellent document, never encodes it, and eighteen months later the distribution is unchanged. The tooling company builds elaborate CPQ approval flows while comp still pays on bookings, and within two quarters reps have reverse-engineered the auto-approval logic and engineer every quote to clear it. The culture company gives inspiring talks about value selling, writes down no specific rule, and discipline lasts until the first competitive deal in week twelve.

They measure headline discount and wonder why margin keeps leaking. Covered above, but it belongs on this list because it is the specific failure the bundling question is asking about. If the governed number is license-line percentage, everything you govern moves to the ungoverned surfaces.

Comp fights the policy, so self-interest wins. Here is the uncomfortable arithmetic. A rep paid on bookings faces a deal that will slip or die without a discount. Discounting is nearly pure upside for them: they get paid on a slightly smaller deal instead of nothing. The margin hit lands on the company's P&L, not the rep's paycheck. So the rep, behaving with complete rationality, treats discount as a cheap tool to de-risk their number. You can write policy against it and add friction, but you are fighting the comp plan on every deal, and self-interest wins the long game.

The fix is to make the discount land in the rep's own paycheck. Margin-based commission pays on gross margin or a margin-adjusted revenue figure, so a deeper concession visibly shrinks the payout. Discount accelerators and decelerators pay a higher rate at or near list and a lower rate on deep discounts, making "hold price" literally more lucrative. Clawbacks recover commission when a deal's discount later proves to violate policy or its economics deteriorate. The mechanism matters less than the principle. And critically, for bundling: the margin calculation must include services and give-aways, or you have simply moved the arbitrage rather than closed it — the rep now protects license margin by pushing every concession into services, and your new comp plan pays them for it.

Leadership exempts itself once. A CRO spends six months rolling out governance, then a marquee deal comes in below band and they override the process personally because it is *their* strategic deal. They experience this as a reasonable exception by the person with authority. The entire sales organization experiences something else: the policy applies to small deals and small people, and when a deal matters it gets set aside — and every rep's deal feels, to that rep, like it matters. One visible exemption teaches the org the policy is optional more powerfully than a hundred slides taught the opposite.

Managers wink at it. Culture is not set at the all-hands; it is set in the one-on-one. Same stalled deal, two managers. Manager A: "What's the value story? Have we quantified their cost of the problem? What's their alternative and why is it worse? Let's build the case before we touch price." Manager B: "Just discount it, I need this in the forecast." Manager B erodes more in one sentence than the all-hands built in an hour. Manager enablement is culture work and belongs in the rollout, not the backlog — train them to coach value defense as a skill, make sure they understand the *why* well enough to defend it under pressure, and measure them partly on their team's margin health.

Only closing carries status. On a bookings leaderboard, a deal closed at list and a deal closed at 40% off look identical, which tells reps that margin is invisible. Counter it deliberately: celebrate the rep who held price with the same visibility as the rep who closed the big logo, create win-at-list recognition as a named category, and run a deal-quality scoreboard alongside bookings ranking margin health, fast-lane percentage, and list-to-effective ratio. Status is the cheapest enforcement mechanism available, because peers enforce it on each other for free.

Governance suspends at quarter-end. The unspoken understanding kicks in, the floor becomes "the floor, but obviously not this week," and the deal desk switches from gatekeeper to expediter. Governance that holds in week three and suspends in week thirteen was never governance; it was a fair-weather process, and discount discipline only matters in the storm. Do not write a quarter-end exception into policy — that codifies the collapse. Build a fast, real escalation path so genuine strategic exceptions move quickly without the floor bending, keep the hard block hard in week thirteen, and have leadership visibly hold the line when it costs them.

They build once and stop. Drift is guaranteed and takes recognizable forms: band creep, where "slightly past" becomes the new normal; exception inflation, until "strategic" means nothing; matrix staleness, as deal-size distribution and product mix move away from the thresholds; approver fatigue; and tooling decay, as new products ship without governance configured. The antidote is a calendared quarterly review owned by RevOps with finance, deal desk, and sales leadership at the table, whose explicit job is to catch drift and re-tighten. Re-tightening is maintenance, not an admission of failure.

Decision framework: when to build what, and when not to build at all

Governance is not always the right investment, and knowing the counter-cases keeps you from building expensive apparatus that solves nothing.

Skip or defer when you are a founder-led team of a handful of reps where the founder sees every deal — formal apparatus is premature overhead, and judgment scales fine at that size. Fix pricing first when the real problem is that list price is wrong; if every deal needs 40% off to close, no governance system will fix that, and the discipline you impose will just cost you deals while you avoid the actual conversation. Reconsider the design when the apparatus itself has become the bottleneck slowing every deal — that is a fast-lane failure, and the fix is band recalibration, not more enforcement. And stop and restart when governance has become a compliance ritual producing clean reports while the discounting behavior is unchanged; reporting that improves while realized price does not is the clearest sign the measured surface and the concession surface have come apart.

Where governance is the right build, the ordering rule is simple: fix whichever leg is missing before strengthening the ones you have. A company with great policy and no tooling gets nothing from a better policy. A company with great tooling and misaligned comp gets nothing from more price rules — it gets more sophisticated circumventing. Diagnose which leg is absent, build that one, and only then tune.

One structural note that decides how well all of this works in practice: the deal desk is the one function touching all three legs simultaneously. It operationalizes policy on live deals and feeds edge cases back into the quarterly review. It runs the tooling and notices when a new product shipped without governance configured. And it is a culture carrier — every interaction between an analyst and a rep is a small culture moment, and a desk that coaches ("here's how to restructure this to stay in band, here's the value argument that lets you hold price") rather than merely polices ("denied") builds the disciplined culture deal by deal. Staff it as a low-level approval-processing function and you get weak policy interpretation, reactive tooling, and no culture at all. Staff it with commercially sophisticated people who understand deal economics and have the standing to hold a line, and it becomes the most valuable operational asset in the system.

Finally, the reporting matters more than it seems, because it is how the program stays funded. The board does not want your discount histogram; it wants the margin story. List-to-effective ratio becomes "our realized pricing is holding." Quarter-end concentration becomes "our discipline survives pressure" — a signal sophisticated board members recognize immediately. Exception volume and fast-lane percentage become "our governance is calibrated." A program that cannot tell that story loses support, then funding, then drifts back to the graveyard it came from.

Related questions

How do you catch free-attach when services aren't in the CPQ?

Track services attach margin per deal in your finance system and join it to the opportunity. If services delivered at or below cost are rising while headline discount is flat, you have free-attach. It is a reporting fix before it is a CPQ fix.

Should the deal desk report to sales or finance?

Either works if the margin floor is finance-owned. Reporting into sales gives the desk commercial credibility and speed; reporting into finance gives it independence. The structural safeguard is not the reporting line — it is that the desk enforces the floor but cannot approve crossing it.

What's a reasonable fast-lane percentage?

There is no universal number, but the direction is clear: most ordinary deals should clear without human approval. A low percentage means bands are miscalibrated or the org routinely operates outside policy — investigate which before adding enforcement.

Do clawbacks damage rep trust?

They do when applied retroactively or ambiguously. Clawbacks work when the triggering conditions are written in the plan before the year starts, tied to policy violations rather than outcomes the rep didn't control, and applied consistently including to top performers.

How does this change for usage-based or consumption pricing?

The concession surface shifts from discount percentage to committed minimums, overage rates, ramp schedules, and credit grants. The same principle holds — govern the total economic concession — but your CPQ has to model committed versus expected consumption to see it.

FAQ

Can you build discount governance without a CPQ?

Partially, and it is worth doing while you procure one. You can write a specific policy, define the total-concession rule, build a manual approval matrix, and — most importantly — start measuring realized price against list from your closed-won data. What you cannot get is the fast lane or the hard block, which means compliance stays voluntary. Treat the manual version as a bridge that produces the business case for the tooling, not as the destination.

What's the single highest-leverage change if I can only do one thing?

Change what you measure from headline license discount to total economic concession, then report it. Most organizations discover a gap of several points between the discount field and reality, and simply making that gap visible starts changing behavior before you have built a single price rule. It also gives you the number that justifies everything else.

How do I get sales leadership to support tighter governance?

Lead with the baseline number and frame it as leverage rather than restriction. A specific policy with a working fast lane makes sales *faster* on the ordinary deal, and it gives sales leadership cover — they can push hard within the guardrails and point at a finance-owned floor as the genuine constraint instead of being the one saying no. The pitch that fails is "we need more control." The pitch that works is "you spend too much of your time on approvals for deals that should never have needed one."

Does margin-based comp make reps avoid low-margin products?

It can, which is why the implementation detail matters. If a strategically important but structurally lower-margin product gets penalized by the comp plan, reps will stop selling it. The usual fixes are product-specific margin targets rather than one company-wide threshold, or a modifier applied against each product's expected margin rather than its absolute margin. Model rep earnings against historical deals before you launch and look specifically for products that get orphaned.

How long before governance actually shows up in the numbers?

Behavioral metrics move fast — quarter-end concentration responds within one quarter, because it is driven almost entirely by whether leadership holds the line at the first crunch after launch. Financial metrics lag two to three quarters, because pipeline quoted under the old rules has to work through. Judge the first quarter on behavior and process compliance, not on realized price.

What do you do when a rep is caught structuring around the rules?

Treat the first instance as a system finding rather than a disciplinary one, because it almost always reveals a hole in the design that other reps are also using. Close the hole, communicate the clarification broadly, and coach the rep. Repeated structuring after a clear, communicated rule is a different conversation — but if you skip straight to discipline on the first case, you lose the diagnostic value and push the behavior further off-system.

Sources

• https://hbr.org/2010/09/how-to-stop-customers-from-fixating-on-price
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/how-b2b-companies-can-win-with-pricing
• https://www.bain.com/insights/pricing-and-revenue-management/
• https://www.gartner.com/en/sales/topics/sales-operations
• https://openviewpartners.com/blog/saas-pricing-strategy/
• https://www.salesforce.com/products/cpq/what-is-cpq/
• https://www.pwc.com/us/en/services/consulting/business-transformation/pricing.html
• https://www.deloitte.com/us/en/services/consulting/services/pricing-profitability-management.html

Related on PULSE

• How do you design a sales compensation plan that rewards margin instead of raw bookings?
• What does a deal desk actually do, and when should you build one?
• How do you structure a CPQ approval matrix that doesn't slow down every deal?
• Why does most of your quarter's discounting happen in the last ten days?
• How do you set a margin floor and get finance and sales to agree on it?
• What pricing metrics belong on a board deck, and which ones mislead?