Quota Capacity Planning for Series B SaaS in 2027

The two models compared, and why you need both

Every Series B capacity plan is really two arguments having a fight. The top-down model starts from the number the CEO sold to the lead investor and works backward to a headcount. The bottoms-up model starts from the roster you actually have — names, hire dates, tenure, segment — and works forward to a capacity dollar figure. Neither one is the plan. The plan is the reconciliation.

Top-down is fast and it is the language the board speaks. You can build a directional version in twenty minutes in a spreadsheet: net-new ARR target divided by (quota × attainment × productive time). Its virtue is that it pressure-tests fundability. If the model says you need 48 AEs to hit the number and your S&M budget supports 31, you have discovered — in twenty minutes, in January — that the plan is not fundable at current quota levels. That is an enormously valuable discovery. Its vice is that it is frictionless in the wrong direction: nothing in the arithmetic stops you from assuming 80% attainment and a four-month ramp, and the model will happily hand you a headcount that presumes top-quartile execution from a team that has never delivered it.

Bottoms-up is slower and it is the language your frontline managers speak. You pull the roster from the HRIS, map every AE to a segment, apply their individual ramp stage, haircut for attrition and non-selling days, and sum. It takes days, not minutes, and it requires you to have clean data on hire dates and territory assignments — which at Series B you frequently do not. Its virtue is that it cannot lie to you about who is actually on the field. Its vice is that it has no opinion about whether the resulting number is acceptable. It just tells you what the current roster can produce.

There is a third model most Series B teams skip and shouldn't: the constraint model, which asks what caps growth other than seller headcount. Pipeline generation capacity is the usual binding constraint. If marketing and SDRs can only source $48M of qualified pipeline and you need 3x coverage on $16M net-new, you are already at the ceiling — hiring a 40th AE adds comp expense and adds zero closed revenue, because the 40th AE has nothing to work. Solutions engineering capacity, implementation bandwidth, and legal/security-review throughput are the other common ceilings. Run the constraint model as a sanity check on both of the other two. If any downstream function is at capacity, your seller-capacity plan is fiction.

The relationship between the three is simple: top-down sets the ambition, bottoms-up sets the floor, and the constraint model sets the ceiling. A defensible plan sits inside all three.

How to decide which number to defend

The decision is not "pick one." It is "which one do you take into the board room as the commitment, and what do you do with the delta." Here is the decision logic RevOps should run before the plan locks.

A few notes on the branches. Pulling hires forward is the cleanest lever because it costs money but not credibility — you are buying capacity, not assuming it. Every 30 days you pull a mid-market hire forward buys roughly a quarter-stage of ramp, which on an $850K quota is real dollars. The constraint is recruiting throughput: if your talent team is already running at capacity, promising eight January starts when they have historically delivered four is just moving the lie upstream.

Raising quota is legitimate only when your current quota sits below the segment benchmark and the raise keeps comp ratio inside the healthy band. If mid-market AEs carry $750K against a $195K OTE, that is roughly a 3.8x quota-to-OTE ratio and you have room. If they carry $1.2M, you do not — pushing higher buys you a spike in voluntary attrition six months later, which destroys more capacity than the quota raise created.

Raising the attainment assumption is the lever most often abused and least often justified. It is defensible exactly once: when you have materially changed something upstream — pipeline coverage moved from 2.2x to 3.4x, you added an SDR pod, you shipped a product that removed a competitive loss reason — and you can show the change in your own data, not in a benchmark report. "We hired a new VP of Marketing" is not evidence.

When the gap exceeds 25%, stop negotiating with the model and escalate. A 25%+ gap means either the target is wrong or the funding is wrong, and both of those are CEO decisions. RevOps' job at that point is to present three funded scenarios — target as-is with more headcount, reduced target at current headcount, and a middle option — with the CAC payback implication of each. That is a far better use of the week than another round of assumption-tuning.

The concrete numbers behind each model

Work a real example end to end. A Series B SaaS company sits at $14M ARR entering the planning year, with a board-approved target of $30M — $16M of net-new ARR, of which roughly $12M is new logo and $4M is expansion. Mid-market is the primary motion.

Top-down. Fully-ramped mid-market quota is $850K. Blended attainment assumption is 70%. Productive time factor is 0.85. The math: $16M ÷ ($850K × 0.70 × 0.85) = $16M ÷ $506K = 31.6 fully-productive AE-years. That is not 31.6 people. It is 31.6 person-years of fully-ramped selling capacity, which is an entirely different thing.

Converting productive AE-years to hired heads. Because new hires ramp and existing reps leave, hired heads always exceed productive AE-years. At Series B mid-market the conversion factor typically runs 1.4x–1.6x. Applying 1.45x: 31.6 × 1.45 = roughly 46 hired AEs on the roster by year-end. If you start the year with 26 AEs, you are hiring 20 net — and given 24% annual attrition on a growing base, that's closer to 28 gross reqs. Recruiting eight AEs a quarter with a Series B talent team is a real operating plan, not a spreadsheet line.

Bottoms-up on the same company. Take a new mid-market AE starting February 1 with an $850K full quota on a 0-25-50-75-100 quarterly ramp:

• Q1 (partial, ramping at 0%): $0
• Q2 at 25% of quarterly quota ($212.5K × 0.25): $53K
• Q3 at 50%: $106K
• Q4 at 75%: $159K
• Year-one ramped capacity: $318K — 37% of full quota

Now multiply that pattern across a roster where a third of the heads are new. Say the roll-up lands at $13.1M of ramp-and-attrition-adjusted capacity against a $16M target. That $2.9M gap — about 18% — is exactly the number that belongs in front of the CRO before the board deck closes, not after Q2 misses.

Segment differences that break blended assumptions. SMB AEs typically carry $550K–$700K quotas, ramp in three to four months, and land attainment in the low-to-mid 60s. Mid-market carries roughly $750K–$950K, ramps in four to six months, and attains in the high 50s. Enterprise carries $1.1M–$1.8M, ramps six to nine months, and attains around the low 50s — the enterprise attainment drag is structural, driven by deal concentration and lumpy timing, and a single blended attainment number will systematically overstate enterprise capacity and understate SMB. Model each segment separately or don't bother modeling.

Attrition, modeled correctly. Voluntary AE attrition at Series B SaaS commonly runs in the low-to-mid twenties percent annually. The critical modeling point is that attrition must be applied per-rep per-month, not as an annual average haircut. A rep lost in February costs ten months of capacity; the same rep lost in November costs one. Modeled monthly, attrition typically erodes 12–18% of nominal capacity — materially more than the naive "multiply by 0.76" approach suggests, because losses cluster after comp-plan issuance and after Q1 misses.

Non-selling time. The 0.85 productive time factor is not a fudge. Count it: roughly 15 days PTO, 8 holidays, 5 sick days, 6 days of SKO/QBR/training, and 6 days of admin and pipeline-review overhead comes to about 40 days out of 260, or 15.4%. Series B teams routinely forget SKO drag specifically, then wonder why Q1 lands 8–12% under model.

Comp inputs that bound the whole exercise. Quota-to-OTE is the hinge between capacity planning and comp planning, and they are genuinely the same exercise run by different owners. A healthy Series B operates around 5x–7x quota-to-OTE. A mid-market AE at roughly $200K OTE — call it $110K base and $90K variable on a 55:45 split — should carry $850K–$1.4M. Below 5x, the company cannot afford the comp expense at realistic attainment. Above 8x, reps read the number as unhittable and leave, which converts a comp problem into a capacity problem. The reciprocal — comp ratio, meaning OTE divided by quota — should sit in the 14–20% band for mid-market and 12–16% for enterprise. Anything above 22% will not survive a CAC payback test, and it is the fastest available signal that a plan is structurally broken regardless of how clean the capacity math looks.

Pipeline coverage as a hard input. Three times coverage entering the quarter is the minimum for a 70% attainment assumption to hold; enterprise needs closer to 4x because of slip risk on large deals. If you enter a quarter below 3x, the honest move is to haircut that quarter's capacity by 15–25% in the model rather than hold the assumption and explain the miss in April.

Implementation and sequencing over 90 days

The plan is not done when the model reconciles. It is done when comp plans are signed, territories are published, and every AE knows their number. Sequencing matters because each step has a dependency you cannot skip.

Days 1–30 — top-down and pressure test. Pull the net-new target from the board deck and split it new-logo versus expansion immediately, because expansion capacity usually belongs to a different team with a different quota structure and folding them together corrupts both. Build the top-down in whatever your finance team already trusts — a planning tool if you have one, a well-structured sheet if you don't. Then do the step most teams skip: pressure-test the implied per-rep productivity against two prior years of actuals. If the plan implies per-rep productivity more than 15% above last year's actual, you need a specific, nameable reason. "The market will improve" is not one.

Days 31–60 — bottoms-up and reconciliation. Pull the roster from the HRIS rather than from a manager's spreadsheet, because manager spreadsheets contain reqs that were never approved and omit reqs that were. Map each AE to a segment, apply their individual ramp stage, and maintain a ramp override list — every AE whose actual ramp differs from the segment default, with a documented reason. Overrides are common and legitimate: a rep hired from a competitor into the same vertical genuinely ramps faster; a backfill inheriting live pipeline and existing account relationships typically ramps 20–30% faster than a net-new territory, so model backfills one ramp stage ahead of default. Conversely, a first-time territory carve-out with no inherited pipeline should be modeled one stage slower. The override list is also your audit trail when someone asks in June why a specific rep's number looked the way it did.

Days 61–90 — lock, issue, go live. Comp plans out by Day 75, rep-by-rep quota conversations done by Day 80, territories published by Day 85, plan live Day 90. The conversations are not a formality — a quota a rep does not believe in is a resignation with a delay fuse, and it is far cheaper to find out in February than in July.

Ownership, explicitly. The CRO owns the integrated number. RevOps owns the model. Finance owns the dollars. The CEO owns the board narrative. All four work from one published source of truth, refreshed within 24 hours of any roster change. When those four disagree in September about what the plan said in January, the plan is already lost.

Ongoing governance. Set a monthly variance review owned by RevOps, attended by CRO and CFO, that compares three things: hires actually made versus planned, ramp progress versus curve, and attainment versus assumption. Each of those has a different remedy. Hiring slippage is a recruiting problem — and slippage is chronic at Series B, where target start dates routinely slip by well over a month, quietly deleting a quarter of capacity from H2. Ramp slippage is an enablement problem. Attainment slippage is a pipeline or product problem. Conflating them produces the classic Series B failure mode: hiring more reps to fix an attainment problem, which raises burn and lowers per-rep productivity simultaneously.

Where capacity planning connects to everything downstream

Quota capacity planning does not end at the AE roster, and treating it as a self-contained exercise is how Series B teams build plans that are internally consistent and operationally impossible.

SDR and pipeline-generation capacity. If AE capacity assumes 3x coverage, someone must build 3x. Back-solve it: $16M net-new at 3x means roughly $48M of qualified pipeline, and if SDR-sourced is meant to be 40% of that, you need about $19M of SDR-sourced pipeline. At a realistic per-SDR annual sourced-pipeline contribution, that dictates SDR headcount as rigidly as the AE math dictates AE headcount. Plan them together or the ratios drift and you end up with expensive AEs doing their own prospecting at a fraction of an SDR's efficiency.

Sales engineering and post-sale. SE-to-AE ratios of roughly 1:3 to 1:4 in mid-market and 1:2 in enterprise are typical, and SEs are usually harder to hire than AEs. Implementation and CS capacity matter for a subtler reason: if onboarding backs up, churn rises, net revenue retention falls, and the net-new number you're planning has to be larger to hit the same ending ARR. Capacity planning that ignores retention is planning gross when the board bought net.

Territory design. Capacity and territory are the same constraint viewed from different angles. A capacity model that says you can support 46 AEs is meaningless if the addressable account universe only supports 34 non-overlapping books at a viable account count per rep. Run an account-coverage check alongside the capacity model: total addressable accounts in segment divided by target accounts per rep gives you a hard headcount ceiling, and if that ceiling is below your capacity-derived headcount, the answer is segment expansion or a new motion, not more reps in the same pond.

Adjacent motions worth modeling separately. Partner-sourced and product-led pipeline behave nothing like AE-sourced and should never sit inside the same capacity assumption. PLG-sourced expansion often needs no AE capacity at all below a deal-size threshold, which is exactly why a blended model that buries it inside AE quota systematically overstates required headcount. Carve them out, quota them separately, and let the AE capacity model cover only what AEs actually close.

The efficiency frame. Every capacity decision at Series B is ultimately a CAC-payback decision. Boards now weigh efficiency alongside growth much more heavily than they did in the 2021 era, and a plan that hits the ARR number while pushing payback past a year is not a win. Before committing to headcount, run the payback math on the marginal rep: fully-loaded cost including OTE, benefits, tooling, and allocated management versus expected first-year gross-profit contribution at modeled attainment. If the marginal rep does not pay back inside your target window, the answer is fewer, better-supported reps — not more reps at a lower quota.

Related questions

What if we don't have clean hire-date data for the bottoms-up model?

Use offer-accept dates as a proxy and flag every estimate. An imperfect bottoms-up model with documented assumptions beats no bottoms-up model. Fix the data pipeline in parallel — HRIS-to-CRM roster sync is a two-week project that pays back every planning cycle after.

Should expansion revenue sit inside the same capacity model as new logo?

No. Expansion typically carries different quota levels, shorter cycles, and often a different team. Model it separately, then combine at the top for the board number. Blending them hides which motion is actually underperforming.

How often should the capacity model be refreshed after Day 90?

Monthly at minimum, plus immediately on any roster change. Quarterly is too slow — a February departure discovered in April has already cost two months of unrecoverable capacity and a delayed backfill req.

What quota-to-OTE ratio signals a plan in trouble?

Below 4x means comp expense will outrun affordability at realistic attainment. Above 8x means reps read the number as unhittable and leave. Series B mid-market should land in the 5x–7x range; drifting outside it is an early warning, not a rounding issue.

Does AI tooling justify raising quotas?

Only with your own cohort-over-cohort evidence. If reps onboarded after your enablement stack shipped demonstrably reach first deal faster than prior cohorts, compress the ramp curve. Absent that data in your own system, leave the assumption alone.

FAQ

How do I reconcile top-down and bottoms-up capacity models?

Build both, compare, and target a gap under 10%. Under 10%, commit to the bottoms-up number. Between 10% and 25%, work the closable levers in order: pull hires forward, raise quota if it sits below benchmark and comp ratio allows, raise attainment only with upstream evidence. Above 25%, escalate — that is a target or funding decision, not a modeling exercise.

What ramp curve should I use for new AEs?

A 0-25-50-75-100% quarterly ramp is the durable default for mid-market. SMB can run faster, roughly 25-75-100-100. Enterprise needs six to nine months, often modeled 0-0-25-50-75-100. Model backfills one stage faster than default because they inherit pipeline; model first-time territory carve-outs one stage slower.

What attainment rate should I assume?

Start at your own trailing twelve-month actual, segmented. If you must use an external anchor, treat 70% as an aggressive-but-defensible mid-market assumption and stress-test the plan at 60% and 80%. A model assuming high attainment is implicitly forecasting top-quartile execution — fine if pipeline coverage and rep tenure support it, dangerous otherwise.

How many AEs do I need for a given net-new ARR target?

Divide net-new ARR by (fully-ramped quota × attainment × productive-time factor) for productive AE-years, then multiply by 1.4x–1.6x to convert to hired heads. Ten million in net-new at $500K quota, 70% attainment, and 0.85 productive time gives roughly 34 productive AE-years — about 48–54 hired heads.

Why is the bottoms-up number always lower?

Because it counts reality: partial-year hires, ramp stages, attrition timing, and non-selling days. Top-down implicitly assumes a fully-ramped, fully-staffed team on January 1, which no Series B company has ever had. If your bottoms-up comes out higher than top-down, you have a modeling error — usually double-counted expansion or a missing ramp haircut.

What's the single most common Series B capacity planning mistake?

Treating hiring plans as hiring outcomes. Target start dates slip routinely, and every slipped month of a mid-market hire deletes real capacity from the back half of the year. Model planned hires with a slippage haircut, and review hires-made-versus-planned monthly so the correction happens in March rather than September.

Sources

• https://www.bridgegroupinc.com/saas-ae-metrics
• https://www.repvue.com/
• https://openviewpartners.com/blog/saas-benchmarks/
• https://www.saastr.com/
• https://www.clari.com/
• https://www.gong.io/resources/
• https://www.bvp.com/atlas/state-of-the-cloud
• https://carta.com/data/
• https://www.pavilion.com/

Related on PULSE

• [Attrition Buffer in Sales Capacity Planning in 2027](/knowledge/ra0424)
• [How to set capacity plans that match Series B headcount budgets in 2027](/knowledge/ra0319)
• [Sales Capacity Model Design for SaaS in 2027](/knowledge/ra0253)
• [How to design rep ramp plans that get AEs to quota in 90 days in 2027](/knowledge/ra0333)
• [Sales Quota Crediting Rules + Edge Cases in 2027](/knowledge/ra0279)
• [Quota Setting: Top-Down vs Bottoms-Up Reconciliation in 2027](/knowledge/ra0252)