What's the right capacity-planning model for a 50-rep org?

The Capacity Formula That Survives a Board Review

A capacity model that survives scrutiny does three things a spreadsheet of quotas cannot: it separates *stated* from *committable* capacity, it makes every discount assumption visible, and it ties each number to a lever leadership can actually pull.

Walk the formula variable by variable.

Quota_i. The individual quota, typically set at 4–5× OTE for a closing AE in mid-market SaaS. A rep on $220K OTE carrying a $900K–$1.1M number sits squarely in that band. If your quotas imply a multiple above 6×, you are either paying below market or setting numbers reps cannot hit — both drive attrition, which feeds back into the model as drag. If the multiple is below 3.5×, your unit economics are probably underwater. The multiple is a diagnostic, not just an input.

Tenure Productivity_i. A factor between 0 and ~1.3 that captures how much of quota a rep at a given tenure actually delivers. A first-quarter closer does not produce like a third-year closer, and a third-year top performer often produces above 100%. This is a *steady-state* multiplier — where a rep lands once fully ramped, given their track record and seat.

Ramp Factor_i. A time-based multiplier for reps still climbing to full productivity. A rep in month two contributes almost nothing; a rep in month eight contributes most of their eventual number. Ramp and tenure productivity are distinct: ramp is *where on the curve they are today*, tenure productivity is *how high their curve tops out*. Conflating them is a frequent modeling error.

Territory Realism_i. A 0.8–1.15 multiplier reflecting patch quality. Two reps with identical skill and identical quotas will not produce identically if one owns a dense, high-ACV patch and the other owns thin, over-farmed accounts. Score every territory and apply the multiplier honestly; this single adjustment can move total effective capacity by 10–15% with zero change to headcount.

Attrition Drag. The whole-team reducer. Voluntary attrition in SaaS sales commonly runs 25–35% annually, and each departure creates a gap: the seat sits empty during recruiting (often 60–90 days) and the replacement then ramps for months. Model this as lost productive days, not just as a headcount subtraction. A team losing a third of its reps a year, each replaced after a 90-day search plus a 6-month ramp, permanently carries several "phantom" seats that appear on the org chart but close nothing.

The output of this formula is what you commit. Suppose the 50 stated quotas sum to $38M. After tenure, ramp, territory, and attrition adjustments, the committable number might be $26M–$28M. That risk-adjusted figure is your board commitment. The $10M+ gap between stated and committable is your headroom for over-attainment, accelerators, and the "stretch" story — not a number you promise. CROs get fired in Q3 for committing stated capacity in Q1.

The Ramp Curve and Tenure Productivity

Ramp is the load-bearing variable most plans get wrong, because it is the one that quietly turns a hiring-heavy year into a missed year. The instinct is to book a new AE at full quota the quarter they start. Reality is a curve, and in mid-market SaaS with multi-stakeholder deals and 60–120 day sales cycles, that curve is long.

A realistic tenure-productivity curve for a mid-market AE looks roughly like this:

The practical consequence is arithmetic, not opinion. A team where 40% of headcount has under 12 months of tenure sits structurally 15–25% below its stated capacity, because a large share of the roster is still on the left side of the curve. This is precisely why aggressive hiring years so often miss plan: leadership counts heads and books quota, the model counts ramped productivity, and the gap between them is the miss.

Two rules follow directly.

First, model ramp at the planning date, not as an annual average. A rep hired in October contributes almost nothing to that fiscal year but nearly full productivity to the next. If you smear ramp across an annual figure, you overstate the current year and understate the following one. Place each rep on the curve by their actual start date.

Second, respect the ramp-capacity of your enablement function. Enablement is a throughput constraint. A single enablement resource can meaningfully onboard only so many reps at once — commonly on the order of a handful per quarter before quality degrades. Hire AEs faster than enablement can ramp them and the curve flattens: months-to-productive stretches from 8–9 to 12–13, and every over-hired rep drags the cohort. The fix is to pace hiring to enablement capacity, or to fund enablement ahead of the hiring wave. Either works; ignoring the constraint does not.

When you present the model, show the tenure distribution of the roster explicitly. A board that sees "50 reps" hears "50 productive reps." A board that sees "28 fully ramped, 15 in ramp, 7 in their first two quarters" understands why committable capacity is well below the sum of quotas — and why the plan is honest rather than sandbagged.

Building the 50-Rep Model Role by Role

Fifty "reps" is never fifty identical closers. A functioning 50-person revenue org is a mix of new-logo AEs, expansion AEs, SDRs, quota-carrying CSMs, sales engineers, and the managers who hold it together. Capacity planning has to model the mix, because the ratios between roles determine whether the closers can actually close.

A representative build for a mid-market SaaS org — call it roughly $300K average ACV — might look like this:

Add the supporting roles and you land near 50 heads, with roughly 36 of them carrying a direct number. The ratios are the substance here:

• SDR-to-AE (~1:2 to 1:3). Too few SDRs and AEs self-source, which cuts their closing time and lowers effective capacity even though headcount looks fine. Too many and you flood the funnel with low-quality meetings that erode conversion.
• Manager span (6–10). Below six, you are over-managing and paying for it in comp overhead. Above ten, coaching collapses, ramp stretches, and attrition rises — which feeds the drag term. Nine is a common target for experienced managers; new managers should carry fewer.
• SE coverage (1:3 to 1:5). In technical mid-market sales, under-resourcing SEs directly lowers win rate on exactly the largest deals, which is a capacity loss that never shows up in a headcount count.

Now convert to capacity. If the direct-carrying roles sum to roughly $38M of stated quota, applying the tenure curve, a ~28% attrition drag, and a realistic ~70% historical attainment yields a risk-adjusted number closer to $26M–$27M. That is the board commitment. The delta to $38M is upside, not plan.

The role-by-role view also surfaces the cheapest capacity levers. Adding two SEs to lift win rate on complex deals, or rebalancing accounts off thin territories, frequently adds more risk-adjusted capacity per dollar than hiring more AEs onto an already-strained ramp. Model the support roles as capacity multipliers, not as overhead, and the hiring plan gets smarter.

Sequencing the Hire Plan and Killing Phantom Capacity

The *order* in which you hire changes how much capacity you actually get from the same headcount budget. A plan that adds the right number of reps in the wrong sequence still misses, because ramp and management are throughput constraints. A few battle-tested sequencing rules:

1. Hire managers ahead of their teams — by 60–90 days. A first-line manager who starts the same week as four new AEs will lose at least one of them inside a year, because nobody was there to onboard, coach, and course-correct during the critical first quarter. Seat the manager a quarter early so the team lands into structure, not chaos.

2. Hire SDRs in pairs, not singletons. Solo SDRs churn hard — the role is high-pressure and isolating, and a lone SDR has no peer to benchmark against or learn from. Pairs cross-train, compete constructively, and cover each other's ramp. Two SDRs hired together produce far more than twice one SDR hired alone.

3. Pace AE hiring to enablement throughput. As above, do not hire AEs faster than enablement can ramp them. If enablement can cleanly onboard a handful per quarter, a wave of ten at once will flatten the whole cohort's curve. Stagger the hires or fund enablement first.

4. Bake backfill lag into the plan. A departing rep is not instantly replaced. Recruiting runs 60–90 days, then the replacement ramps for months. Apply an effective-FTE multiplier — often around 0.75 — to any role with meaningful turnover, so the plan reflects seats that are open or ramping rather than closing.

This is where phantom capacity dies. Phantom capacity is the gap between the org chart (50 heads) and the closing reality (perhaps 60–70% of that number producing at any given moment). One useful mental model is a rough split: a majority of the roster fully productive, a meaningful slice in ramp, and a small slice in active backfill at all times. Plan against the productive slice, not the headcount, and you stop over-committing.

The discipline that prevents over-leveraging is a single comparison: stated capacity versus committable capacity. Build the bottom-up sheet with three columns per rep — stated quota, tenure-and-ramp-adjusted productivity, and "what we'd commit to the CFO." If the gap between column one and column three exceeds ~25%, the headcount plan is over-leveraged. The right move is not to hire faster; it is to *pause* hiring until enablement, territory, and management span are repaired. Adding reps onto a broken ramp curve burns cash and morale at the same time, and it makes the miss worse, not better.

The Bear Case, Territory Realism, and Monthly Health Checks

A capacity model that only shows the base case is a liability. The CFO will run the pessimistic version whether you present it or not, and it is far better to bring it yourself with the levers already attached.

Run the bear case explicitly. Take the base assumptions — say 70% attainment, 28% attrition, 9-month ramp — and replay the model with the dark scenario: attainment at the median rather than the goal (often closer to 55%), attrition at 38% (what actually happens when comp is cut or a competitor poaches a pod), and ramp stretched to 13 months (real-world when enablement is under-resourced and deals are complex). Under those inputs, risk-adjusted capacity can fall by 30–40% versus the base plan. That is not paranoia; it is what a single quarter looks like when a top performer leaves, a territory rebalance disrupts pipeline, or the macro tightens and deals slip. The credible plan shows base / base-minus-bear / bear, and for each scenario names the trigger metric and the response — for example, "if rolling 90-day attainment falls below 60%, freeze hiring and reallocate accounts." CROs who present only the base case get blindsided; CROs who present the range and the triggers earn a longer runway.

Score territory realism with a real scorecard. Do not apply a flat territory multiplier. Build a 1–5 density score per patch from three factors: account concentration (count of genuinely qualified high-ACV accounts in the territory), industry mix (diversified patches are more resilient than single-sector ones), and recent pipeline velocity (weighted pipeline-to-quota over the last two quarters). In a typical 50-rep org you will find a small band of high-density patches (score 4–5), a large middle (score 3), and a meaningful tail of thin patches (score 1–2). Apply roughly a 0.8× multiplier to thin territories and 1.15× to dense ones in the capacity formula. This surfaces which territories need investment or consolidation *before* you hire into them — and it often reveals that fixing coverage adds more capacity than adding heads.

Run the health check monthly, not quarterly. A 50-rep org moves fast enough that a quarterly cadence misses the 3–4 month lag between a hiring decision and its revenue impact — a lag that silently erodes plan. Use a tight dashboard of four numbers:

• Rolling 90-day attainment by tenure cohort. If the under-12-month cohort is below ~50% by month nine, the ramp design is broken and no amount of hiring fixes it.
• ARR-per-FTE trend. Should hold in a stable band for your segment; a falling trend signals over-hiring relative to demand.
• Pipeline coverage by stage and rep. Roughly 3× for a 90-day forecast is the floor; push toward 4× when win rates run below 25%.
• Backfill gap days. Average days between a departure and the replacement reaching ~50% productivity. When this exceeds ~75 days, halt new hiring until the backfill pipeline clears — you are opening seats faster than you can fill them.

Attach a threshold and an action to each metric, and the model stops being a once-a-year artifact and becomes an operating instrument. The single most important habit: keep the bear-case number and its triggers visible in every monthly review, so the org is never surprised by a scenario it already modeled.

FAQ

What's the first step in capacity planning for a 50-rep org?

Build bottom-up unit math per rep before you touch a top-down headcount or revenue target. Compute each rep's contribution as quota × tenure productivity × ramp factor × territory realism, then sum. Only after you have that bottom-up figure should you compare it to benchmarks. Starting top-down — taking a revenue goal and dividing by an average quota — bakes in an optimism that the bottom-up view would have caught.

How do I sanity-check my bottom-up capacity model?

Compare the total against three independent top-down anchors: ARR-per-rep of roughly $700K–$1.1M for mid-market SaaS, individual quota at about 4–5× OTE, and manager span-of-control of 6–10 ICs per first-line manager. If the bottom-up build and these anchors disagree by more than about 15%, there is a hidden assumption in the model — most often an over-optimistic ramp curve or stale territory data. Find and fix it before presenting.

How much does attrition really change the number?

A lot, and it is the variable most plans underweight. SaaS sales attrition commonly runs 25–35% annually, and each departure costs a 60–90 day recruiting gap plus a multi-month ramp for the replacement. Model it as lost productive days, not a simple headcount subtraction. Ignoring attrition and backfill lag typically overstates real capacity by 20–30%, which is exactly the size of a plan miss.

How should I factor in ramp time for new hires?

Place each rep on a tenure curve by their actual start date rather than averaging ramp across the year. In mid-market SaaS, expect roughly 0–15% attainment in months 0–3, 25–50% in months 4–6, 60–80% in months 7–9, and full productivity by months 10–12. A rep hired in the fourth quarter contributes little to that fiscal year and most of their value to the next — so smearing ramp into an annual average distorts both years.

What's a realistic territory realism adjustment?

Territory realism captures how much a patch's quality helps or hurts an otherwise-identical rep. Score each territory 1–5 on account concentration, industry mix, and recent pipeline velocity, then apply roughly a 0.8× multiplier to thin patches and up to 1.15× to dense ones. Across a 50-rep org the blended adjustment usually lands around 0.85–0.95, depending on data quality and coverage — and rebalancing thin territories can add 10–15% of effective capacity with no new headcount.

How often should I update the capacity model?

Monthly for an org this size. A quarterly cadence misses the 3–4 month lag between hiring decisions and revenue impact, which quietly erodes plan. Refresh a rolling 12-month view each month, watch attainment by tenure cohort, ARR-per-FTE, pipeline coverage, and backfill gap days, and re-run the bear case any time you lose a top performer, rebalance territories, or change comp. Annual-only planning is how orgs discover a miss a quarter too late to fix it.

Sources

• Bridge Group — annual SaaS AE and SDR metrics and compensation research: https://www.bridgegroupinc.com
• Gartner — sales organization design, quota-setting, and capacity research: https://www.gartner.com/en/sales
• Harvard Business Review — frameworks on scaling sales teams and sales-force sizing: https://hbr.org
• McKinsey & Company — sales-force effectiveness and go-to-market resource allocation: https://www.mckinsey.com
• Pavilion — State of Revenue Operations and go-to-market benchmarks: https://www.joinpavilion.com
• Xactly — sales performance and compensation benchmark research: https://www.xactlycorp.com

Related on PULSE

• [How do you build a real bottom-up forecast in a 50-rep SaaS org that doesn't fall apart when one AE has a $2M deal slip?](/knowledge/q9517)
• [How do I build a real bottom-up forecast in a 50-rep org?](/knowledge/q49)
• [Top-down vs. bottom-up quota models — and when each is the right call](/knowledge/q729)
• [How to scale a sales team from 10 to 30 reps in 9 months without crushing win rate](/knowledge/q1112)
• [Territory design for a 30-rep mid-market team](/knowledge/q116)
• [Realistic AE ramp timelines in mid-market SaaS](/knowledge/q17)