One Big Table Consulting ← All Advanced Analytics

Domain 3 — Attainment curve design and payout mechanics, dark OBT theme

Domain 3 — Attainment curves and payout mechanics
Core design layer · 6 sections · Live curve builder · Quiz at end
What an attainment curve actually is
The attainment curve is the function that converts a rep's performance percentage into a multiplier on their TI. It is the single most important behavioral lever in plan design. Every rep decision about where to invest their time — which deals to pursue, how hard to push in Q4, whether to accelerate or hold pipeline — is a rational response to the shape of this curve.
Incentive = TI × curve(attain%)
Total payout = Base + Incentive
curve(attain%) returns a multiplier: 0 below threshold, ramps to 1.0 at quota, accelerates above quota
The curve is a piecewise function — different rules apply in different attainment zones. Understanding each zone and its behavioral implication is the core analytical skill this role requires.
The four zones of every attainment curve
Zone 1 — Below threshold (0% → threshold%)
curve = 0. No incentive paid regardless of sales. Threshold is typically 0–50% of quota. Behavioral signal: reps below threshold have no marginal incentive until they cross it — creating a "dead zone" where effort doesn't pay. Setting threshold too high (e.g. 70%) risks leaving demotivated reps with no path to recovery mid-year.
Zone 2 — Linear ramp (threshold% → 100%)
curve ramps linearly from 0 to 1.0. Every incremental attainment point earns the same marginal payout. Behavioral signal: equal incentive to improve at every level — good for motivating the middle of the distribution. The slope of this ramp = TI / (quota × (1 − threshold%)).
Zone 3 — Accelerator (100% → cap%)
curve exceeds 1.0 and accelerates at a rate above the linear zone. Behavioral signal: reps earn disproportionately more for performance above quota — the primary driver of "hunting" behavior. The steeper the accelerator, the stronger the pull. Also the primary source of cost overrun risk.
Zone 4 — Cap (above cap%)
curve is fixed at cap value. No additional incentive above this point. Behavioral signal: reps who hit the cap have no incentive to keep selling — they may bank pipeline for next period (sandbagging). Setting cap too low kills top performer motivation. Setting it too high creates unlimited cost exposure.
Dead zone
Linear ramp
Accelerator
Cap
Why this matters for your role
The comp analytics team's job is to model the cost and behavioral consequences of each zone before the plan goes live. Changing the threshold from 50% to 0% increases cost in downside scenarios. Changing the accelerator from 150% to 200% increases P90 cost significantly. Every zone parameter change must be modeled — that is this role's core deliverable.
The four curve types — design choices and tradeoffs
Linear curve
ShapeStraight line, constant slope
Formulacurve = attain% / 100
Best forSimple transactional roles
RiskNo behavioral pull at quota — reps have equal incentive at 80% and 100%
Cost profilePredictable — scales linearly
Accelerated curve
ShapeKink at quota, steeper above
FormulaLinear below, accel × rate above
Best forEnterprise / quota-carrying AEs
RiskCost tail risk if many reps land in accelerator
Cost profileVariable — tail-heavy in upside scenario
Stepped / tiered curve
ShapeFlat then jumps at defined tiers
FormulaDiscrete payout multiplier per tier
Best forRoles where exact attainment % is harder to measure
RiskGaming — reps hold pipeline to hit next tier. Cliff effect.
Cost profileLumpy — spikes at tier boundaries
Exponential / convex curve
ShapeAccelerates continuously — steeper as attainment rises
Formulacurve = attain^n (n > 1)
Best forStrategic / hunter roles with high deal variance
RiskExtreme cost tail — single large deal can generate massive payout
Cost profileHighly variable — windfall provision required
The curve type selection framework
Three questions determine the right curve type:

1. How predictable is deal size? High variance (enterprise) → accelerated or exponential to reward large deals. Low variance (transactional) → linear.

2. How much cost risk can the company absorb? Tight margin business → linear or stepped with hard cap. High-margin SaaS → accelerated with generous cap.

3. What behavior are you trying to drive? Push past quota → steep accelerator. Consistent activity → linear. Hit specific tiers → stepped.
Say this in the interview
"Curve type is a strategic choice, not a default. I'd always start by asking what behavior the business is trying to incentivize — because the curve literally encodes the answer. An accelerated curve says 'we will pay you disproportionately to outperform.' A stepped curve says 'we care about discrete milestones.' A linear curve says 'every dollar of revenue is equally valuable to us.' The analytics team's job is to model the cost and behavioral consequences of each option before a choice is locked."
Live curve builder — design and see the consequences
Threshold (floor %)50%
Accelerator rate above quota150%
Cap (max TI multiplier %)250%
TI ($K)$60K
Your attainment (%)110%
Incentive earned ($K) Your position Zone fills
Overlays, multipliers, and SPIFFs — the modifier layer
The base attainment curve handles primary quota performance. Overlays and SPIFFs are additional incentive mechanisms layered on top to drive specific behaviors the base curve doesn't address. They are the most common source of unexpected cost spikes — and the most commonly under-modeled component.
The modifier types
Product mix overlay
A multiplier applied to the base payout based on what the rep sold, not just how much. Example: if Azure revenue ≥ 30% of total, apply 1.2× multiplier to full payout. Purpose: prevent reps from optimizing toward easy legacy products at the expense of strategic cloud products.

Cost modeling challenge: the multiplier compounds with the base curve — a rep in the accelerator zone who also hits the mix target earns (base payout × accel multiplier × mix multiplier). These compound dramatically at high attainment.
Team / overlay credit
Multiple reps share credit for one deal. AE + solution specialist + industry overlay each receive a portion of the credit. Split percentages must sum to ≤100% or company double-pays. Behavioral purpose: incentivize collaboration on complex enterprise deals without creating channel conflict.

Analytics flag: validate that all crediting rules are calculable in the comp system (SAP Commissions) before launch — manual split credit is a dispute factory.
SPIFF — Short-term Performance Incentive
A temporary, time-bounded bonus for a specific behavior — launching a new product, closing deals in a specific industry vertical, hitting Q4 pipeline targets. Paid on top of the base plan. Cost modeling challenge: SPIFFs are often approved quickly and modeled loosely — then generate cost surprises because nobody modeled the interaction with the base accelerator. A rep in the accelerator who hits a SPIFF earns: base payout + SPIFF flat amount + potentially SPIFF amount also eligible for accelerator treatment if not capped separately.
Clawback and draw mechanics
Clawback: company reclaims paid incentive if deal cancels or rep leaves within a window (typically 6–12 months). Enforceability varies by state — California is most restrictive, requiring specific language in the plan document.

Draw against commission: advance payment to a new rep during ramp period, recouped from future earned commission. Recoverable draw must be repaid if unearned at separation. Non-recoverable draw is forgiven — essentially a salary supplement. The analytics team must model expected draw recovery rates when forecasting new hire comp cost.
Compound payout example — base + mix overlay
Base attainment (%)130%
TI ($K)$60K
Mix overlay multiplier1.20×
SPIFF flat bonus ($K)$10K
Say this in the interview
"The modifier layer is where plan cost most commonly surprises finance. The base attainment curve is well-modeled. But when a product mix overlay compounds with an accelerator payout, and then a SPIFF lands on top, the total payout for a high performer can be 3–4× their TI — which nobody planned for. My approach is always to model modifier interactions explicitly, especially at the 90th percentile of attainment, before any overlay is approved."
How Microsoft structures attainment curves in practice
The standard Microsoft accelerator ladder
Microsoft uses a stepped accelerator structure rather than a single accelerator rate. A typical enterprise seller plan:

0–50%: 0× (threshold zone, no incentive)
50–100%: linear ramp to 1.0× TI
100–120%: 1.5× rate (accelerator step 1)
120–150%: 2.0× rate (accelerator step 2)
150%+: 2.5× rate (accelerator step 3, capped at 3.0× TI)

Each step creates a behavioral inflection point — reps push specifically to hit 120% and 150% because the marginal payout jumps. The analytics team models the cost impact at each step boundary separately.
Multi-measure plan payout calculation
Most Microsoft plans have 2–4 measures. Total payout is a weighted sum:

Total incentive = Σ (TI_measure × weight_measure × curve_measure(attain_measure%))

Example: Azure consumption (50% weight) + M365 seats (30% weight) + Customer success score (20% weight). Each measure has its own attainment curve. A rep can overachieve on one measure and underachieve on another — the weighted sum determines the final payout.
The kicker / accelerator modifier
Above the primary accelerator, Microsoft sometimes applies a "kicker" — an additional bonus paid when a rep achieves both high overall attainment AND hits the product mix target. This is a compound modifier: base curve payout × accelerator × kicker multiplier. The analytics team must model this three-way interaction explicitly — it is the most common source of cost overruns in Microsoft's comp program.
Decelerator provisions
A less-discussed feature: some Microsoft plans include a decelerator — a payout rate that is lower than 1× below a certain attainment band (e.g., below 70%, you earn at 0.5× rate rather than 0%). This prevents a complete cliff at threshold while still penalizing underperformance. The analytics team models decelerators separately because they change the cost floor in downside scenarios.
Cap mechanics and windfall provisions
Microsoft caps total incentive payout at 3–4× TI for most plans. For large enterprise deals that would trigger windfall payouts (single deal > 40% of annual quota), a windfall review process applies — the deal is reviewed by finance and HR before credit is fully awarded. The analytics team provides the model output showing what the uncapped payout would be and the recommended windfall adjustment.
The interview question this section answers
"Walk me through how you would model the cost impact of changing the accelerator rate from 150% to 200% for the enterprise segment." Answer: run scenario analysis on the attainment distribution for that segment, calculate the payout delta for every rep above 100% under both rates, aggregate to total cost delta, run Monte Carlo on the uncertainty in how many reps will land above 100%. Deliver P50 and P90 cost impact to finance.
Say this in the interview
"The stepped accelerator structure at Microsoft is analytically interesting because each step boundary creates a behavioral inflection. Reps rationally make decisions about pipeline timing around those boundaries — pushing to hit 120% before quarter end because the marginal dollar just got 33% more valuable. That behavioral response is real and measurable in attainment data. Part of my job would be to detect those patterns and flag where the steps are creating unintended pipeline concentration effects."
Domain 3 quiz — seven questions