The quota is the most important number in a revenue organisation. It drives territory design, compensation, headcount, and forecast. It determines who gets paid and who gets managed out. It shapes the culture of a sales floor more than any values statement ever will.
It is also, in most organisations, a fiction.
Not a deliberate fiction — nobody sits in a room and invents numbers from nowhere. But a convenient one. The quota is usually the number that makes the annual plan work, back-calculated from a board-level revenue target and divided among the sales team by headcount or gut feel. The question nobody asks is whether the number is actually achievable by the people being asked to hit it.
That's the problem. And it's not a data problem. It's a process problem.
How most quotas get set
The standard process goes like this. The CFO tables a revenue target in October. The VP of Sales translates it into a per-rep number by dividing by headcount. Someone adds a layer of "stretch" — typically 10–20% above the base plan — on the rationale that if you aim higher you'll land where you need to be. The number is filed, blessed by the board, and announced at the SKO in January with a motivational slide deck and a lot of applause.
This feels rational because the arithmetic works. Twelve reps at £500k each gets you to £6m. If you need £6m, you have your quota. But the question being answered is not "what can these reps hit?" It is "what number makes the spreadsheet balance?" These are different questions. Most organisations never ask the second one.
The result is quotas that share three characteristics: they are identical across reps regardless of territory size, ramp stage, or product mix; they rise in proportion to the top-line target rather than any change in market opportunity; and they are presented to new hires as a contractual obligation before those hires have closed their first deal.
A rep whose territory has 40 addressable accounts with an average deal size of £25k has a ceiling of £1m — before you account for ramp, cycle length, or the 30% of their time they spend in internal meetings. Give that rep a £750k quota and you have created a structural problem. Call it an accountability problem, a performance problem, or a culture problem, but the origin is in the number, not the person carrying it.
The three questions a quota needs to answer
Before a quota number is defensible, it has to survive three questions. Most quotas cannot answer any of them.
1. Is the market big enough?
The first question is about ceiling. Every rep operates in a defined territory — a set of named accounts, a vertical, a geography, or some combination. That territory has a total addressable value: the number of accounts that could plausibly buy, multiplied by the average deal size for accounts of that type.
If that ceiling is less than three times the rep's quota, you have a lottery, not a plan. At 3× coverage, the rep has to close one in three addressable opportunities — which is an aggressive close rate even in a market the rep knows well and a product the market genuinely wants. Below 3× coverage, the quota is arithmetically implausible and the rep knows it by end of Q1.
Most organisations do not calculate this number before setting quota. They assign territories after the quota is set, which reverses the causal logic entirely. The coverage ratio should constrain the quota, not the other way around.
2. Do we have the capacity to hit it?
The second question is about throughput. A rep is not a quota machine. They have a sales cycle that limits how many deals they can close per year and a deal load that limits how many they can carry simultaneously. Both of these are knowable from historical data — not modelled, not assumed, actually measured from closed-won records.
Take a rep with a 90-day average sales cycle. If they can carry 12 active deals simultaneously without quality degrading, they can progress roughly 48 deal-slots per year — four full cycles of 12. Apply the real close rate (not the CRM close rate, which typically includes deals that were never real; the actual rate from deals that genuinely entered the sales process) and you have a throughput number. Multiply by average deal size. That is the capacity ceiling.
Quota should sit at 80–90% of capacity. Not at 110%. Not at "stretch." At a level the rep can reach in a year where nothing catastrophic happens, without requiring a miraculous Q4 to compensate for a structurally broken H1.
3. Does history support it?
The third question is the simplest and the least comfortable. Last year, what did your top 25% of reps hit? What did the median rep hit? What percentage of reps hit quota at all?
If your quota sits above what 75% of reps achieved last year, you do not have a quota problem. You have a hiring problem, a product problem, or a territory problem — and no quota setting methodology will fix any of those. The quota is being asked to carry the weight of something it was never designed to carry.
The attainment distribution is the most honest read on quota health available. More than 20% of reps hitting 150%+ of quota means the quota is too low. Fewer than 50% hitting quota — consistently, not as a Q4 aberration — means the quota is too high or the inputs are wrong. The distribution does not lie. The rationalisation of it almost always does.
The bottom-up build
A quota built from the bottom up starts not with a revenue target but with the evidence of what a rep in this organisation, in this market, with this product, can actually close in a 12-month period.
Step one: establish real average deal size. Pull it from closed-won data, exclude the top 5% of outliers (the enterprise anomaly that distorts the mean), and calculate separately by segment if your AEs cover multiple customer tiers. The number you want is the one a median rep can realistically repeat, not the one that looks best in a board presentation.
Step two: establish the real close rate. Not the rate your CRM reports, which typically includes pipeline created by SDRs for deals that were never real, deals that went dark after the first call, and re-heated deals from previous quarters that somebody re-staged to hit an activity metric. Pull the real close rate: deals that entered a genuine sales process (discovery completed, decision-maker confirmed, budget established) divided by deals that closed won. In most B2B SaaS organisations this number is 20–35%. In many organisations the CRM reports something closer to 45%. The gap is where optimism lives.
Step three: factor ramp. A new hire at month one does not have 12 months of selling capacity. They have whatever the ramp schedule allows — typically full productivity by month four to six in a complex sale. A new hire hired in August and ramped by February has contributed meaningfully to one annual quota cycle. Treat them as such.
Step four: subtract non-selling time. Every rep spends time on internal meetings, admin, CRM hygiene, training, and QBRs. In most organisations this accounts for 25–35% of available working hours. A rep with a 250-day selling year is actually working approximately 160–180 selling days. The capacity model needs to reflect this, because the rep's calendar already does.
The output of these four steps is a capacity ceiling. The quota should be 80–90% of that ceiling. This gives the organisation a plan it can defend and the rep a target that rewards genuine performance rather than punishing structural constraints disguised as personal failure.
The attainment distribution validates the number after the fact. If it is healthy — roughly bell-shaped, centred around 90–110% attainment, with a small right tail of overachievers — the quota was calibrated correctly. If the distribution is bimodal (a cluster at 60% and a cluster at 140% with nothing in between), the territory design is broken and the quota is answering the wrong question.
What top-down quotas actually cost
The obvious cost is missed targets. The less obvious costs are where the real damage accumulates.
Sandbagging is the first casualty of a top-down quota that has been set too high. Reps who discover, usually by Q2, that their quota is structurally unachievable do not respond by working harder. They respond by protecting their baseline. They slow down in Q4 to avoid setting a precedent. They hold deals across the quarter boundary. They manage up rather than managing their pipeline. The organisation reads this as a performance issue. It is a quota design issue.
End-of-quarter discounting is the second. When reps cannot hit quota through volume, they compensate with concessions. The deal closes — technically — at 30% below list price, with a 12-month payment plan and a first-year success guarantee. The revenue is recognised. The unit economics are quietly destroyed. Finance attributes this to sales discipline. RevOps should attribute it to quota pressure on a pipeline that was never going to carry the weight.
The third and most expensive cost is attrition. Your best reps are the most employable. They are also the most capable of recognising that the quota is structurally unachievable — because they have the market knowledge to understand their territory ceiling and the pipeline discipline to know when the maths does not work. They leave before the organisation has processed why. Replacement cost for a fully ramped enterprise AE is typically 1.5–2× annual OTE. That cost is almost never attributed to quota design. It should be.
The Framework
The full interrogation framework for this is Dispatch #003 — The Quota. 38 questions across four sections: Assumption Exposure, Bottom-Up Construction, Capacity Coverage, and SKO Sanity Check. $97. Instant download.
See the full framework →The SKO problem
The SKO is the wrong place to have the quota conversation. By January, the number has been through board approval, territory allocations have been built on it, OTE letters have gone out, and changing it requires a political fight that most RevOps leaders do not have the standing to win. The conversation that happens in January is a performance conversation dressed up as a planning conversation. The decisions are already made.
The right time to have the quota conversation is October — before the board approves the revenue target, while there is still room to model attainment scenarios and present them credibly to the CFO. This requires a RevOps function that can run the bottom-up build and show the gap between what the revenue plan requires and what the organisation's current capacity can deliver. It requires leadership that treats the attainment model as evidence rather than as negotiating theatre.
Most RevOps functions do not get the October meeting. They get a January briefing and a request to "build a plan to hit the number." The consequence is that the quota conversation happens reactively — in Q2, when attainment is behind, when the data is already damning, and when the options are limited to discounting the pipeline or discounting the people who built it.
The SKO sanity check — a structured review of the quota assumptions in the week before announcement — is the last line of defence for a number that was set without the benefit of the bottom-up build. It does not fix the structural problem. But it surfaces the most dangerous assumptions before they are committed to 12 months of OTE letters and territory contracts. A quota that has been examined is at least a quota that has been asked the right questions.
Common objections and why they're wrong
Quota reform conversations attract a predictable set of objections. They are worth addressing directly, because they tend to appear in planning meetings dressed as conventional wisdom rather than as what they actually are: rationalisations for not doing the harder work.
"If we lower the quota, people will just do less."
This is the most common objection and the least supported by evidence. The behaviour it describes — reps easing off once they hit target — is a symptom of quota being too high, not too low. Sandbagging happens when reps have learned through experience that overachievement leads to a higher quota next year. The rational response to that incentive structure is to protect the baseline. A rep at 80% of an impossible quota is not more motivated than a rep at 100% of a realistic one. They are less motivated, more cynical, and more likely to be in a competing AE's Slack next quarter.
"We need stretch targets."
Stretch targets and base quota are different instruments. A base quota is what the organisation plans around — it drives territory design, headcount, and forecast. It should be realistic by construction. A stretch target is what the compensation plan rewards above the base. You build the base from evidence and you build the accelerator above it. Conflating the two produces a base quota that is actually a stretch target, which means the organisation's plan is built on a number it never genuinely expected to hit, which means the forecast is fictional from the first day of Q1.
"Our top reps always hit their quota."
They do. And the reason is almost always that they have the best territories. The rep who covers the financial services vertical in London with 60 named enterprise accounts and a strong referral network will hit their number in most years — not because the quota is well-calibrated, but because the territory is rich enough to absorb a badly calibrated quota without showing the damage. The rep covering mid-market retail in a region of 20 accounts will not hit the same number and will be managed out for a territory design failure that has been relabelled as a performance problem. The top reps masking the issue is not evidence that the quota system works. It is evidence that territory design is doing the work the quota methodology should be doing.
Where to start
The audit comes before the rebuild. Before you change the number, you need to understand the current one. What assumptions built it? What data was actually used, and what data was available but ignored? Where did the top-down number and the bottom-up number diverge, and which one won?
The questions a quota has never been asked are more diagnostic than any attainment report. An attainment report tells you what happened. The assumption audit tells you why — and whether the problem is structural or situational. Structural problems require a different response to situational ones, and you cannot tell the difference without doing the interrogation first.
The place to start is with the current year's quota construction logic. Not the presentation that accompanied the SKO announcement — the actual working assumptions behind the number. What close rate was assumed? What deal size? What ramp timeline? What territory coverage ratio? If those numbers cannot be produced, that is itself a finding. A quota built without documented assumptions is a quota that cannot be defended, adjusted, or learned from. It is a number in a box with no address on it.
The rebuild follows from the audit. Not before.