Do you need more reps, or better ones?

The number came in short. The board asks what you are doing about it. The answer that arrives fastest, in almost every company I have worked in, is that we need more reps.

Sometimes that is correct. More often it is the most expensive way to avoid a diagnosis.

Here is the context that should frame the decision. Average B2B quota attainment has fallen to around 43 percent, down from 53 percent in early 2022. The slope is remarkably consistent: 52 percent in 2024, 46 percent in 2025, 43 percent by the middle of 2026. Roughly 69 percent of B2B reps are now missing their number.

When most of a population fails to clear a bar, the honest conclusion is not that the population got worse.

Start with whether the number was ever real

Before you decide the team is short, check whether the target was built to be hit.

THE MODEL IS MOST USEFUL WHEN IT DISAGREES WITH THE INSTINCT.

Around 58 percent of companies intentionally over-assign quota, typically by 20 to 30 percent. This is not a scandal, it is standard practice, and the logic sounds fine on a spreadsheet. Set the bar high and people stretch. Assign 130 percent of plan across the team and you have covered yourself for slippage.

Except when every rep is carrying an over-assigned number, you have not built a stretch culture. You have built a system where the majority is designed to miss before the quarter opens. Forrester has made the point bluntly: an average attainment near 47 to 50 percent is not necessarily underperformance. It can simply be the arithmetic of how the plan was written.

A team where 70 percent of reps hit a credible number is healthier than a team where 40 percent hit an inflated one. The first retains people and forecasts accurately. The second churns its best sellers, who leave precisely because they can do the maths.

So the first question is not how many reps you need. It is whether the number you are measuring them against was ever achievable with the pipeline they were given.

Then check whether it works for anybody

The single most useful distinction is between a team where nobody hits and a team where two people hit.

If nobody hits, the problem is upstream of the reps. Targeting, motion, pricing, product fit, or a quota that was never real. Adding people to that multiplies the cost without changing the outcome.

If two people hit and the rest do not, you have proof the thing is sellable. What you do not have is a system, and that is a different project entirely.

The number that makes this concrete: in one large analysis, 17 percent of reps generated 81 percent of revenue. That is not a performance distribution. That is a business running on a handful of people who have each privately worked out how to sell it, and none of whom have written it down.

Hiring into that pattern does not help. You are adding people to an organisation that cannot teach, which is why the same analysis found turnover climbing from 22 percent to 36 percent. The new hires arrive, get no transferable method, underperform, and leave. Then you hire again.

The arithmetic that actually answers it

Capacity planning is not complicated, and almost nobody does it before hiring.

Work backwards. Take the number. Divide by average deal size to get deals required. Divide by win rate to get opportunities required. Divide by opportunity-to-meeting conversion to get meetings required. Divide by what one rep can actually hold in a month, accounting for cycle length.

That gives you a headcount figure. Then do the honest part.

Compare it to what your current team is producing per head. If a productive rep is running at 60 percent of the figure the model says one rep should produce, adding a body does not close the gap. It gives you two people at 60 percent and a larger cost base.

The model is most useful when it disagrees with the instinct. When the maths says you need three more reps and the instinct says one, you have been quietly under-resourced for a year. When the maths says you need none and the instinct says two, you have a productivity problem wearing a capacity costume.

What a new hire actually costs

The hire decision gets made as though the cost is the salary. It is not.

Average time to full productivity is around ten months. Structured onboarding cuts that by about a third, to six or seven, and most companies do not have structured onboarding. So the real cost is the base, plus the territory sitting idle, plus the manager hours spent answering the same questions nobody wrote down, plus the accounts burned by somebody learning on live deals.

And you find out whether it worked in month seven or eight, which means a mishire costs close to a year.

That is the number to weigh against fixing what you have. A quarter spent writing down what your two performers do differently is cheaper than a year spent discovering that a new hire could not reverse-engineer it either.

When hiring is the right answer

It is the right answer when the system exists and the capacity does not.

Concretely: your win rate is stable across reps rather than concentrated in two of them. Your ramp is written down and somebody has completed it. A new person can be told what good looks like in their first week without a senior rep improvising it. And the capacity model says you are genuinely short.

Under those conditions a hire compounds, because they arrive into something that teaches. Without them a hire is a bet that the next person will work out on their own what the current team worked out on their own, which is not a strategy. It is a run of luck you are hoping continues.

The order that matters

Check the quota was real. Check whether it works for anybody. Run the capacity maths. Then decide.

Most companies do the last step first and spend the following year explaining a miss that was arithmetic from the beginning.