The most expensive place to practice a sales conversation is on a live call with a real customer. Every sales organization knows this and almost all of them do it anyway, because the cost never appears on a line item.
This is an argument rather than a how-to. The claim is that the training budget is the small number, and the pipeline consumed by reps learning on live deals is the large one.
Where the cost actually sits
Sales training gets budgeted as a spend on content and events. That framing hides the real expenditure, which is paid in pipeline, in ramp months, and in manager time, none of which are tracked as training costs.
Three buckets carry almost all of it, and none appear in an enablement budget.
Bucket one: the deals lost to a first attempt
Every time a rep tries new messaging, a new objection response, or a new product story for the first time on a live call, the buyer is paying for the rehearsal. Sometimes it goes fine. Sometimes the deal ends at that objection.
A HubSpot survey found that around 44 percent of deals fall apart because objections are not handled well, and that only about a quarter of reps feel confident in how they respond. Those two figures describe the same problem from both ends: reps are underprepared for the moment that decides the deal.
Bucket two: the ramp months
Ramp is the largest single training cost in most sales organizations and it is rarely counted as one. The Bridge Group's research on inside sales metrics has consistently put average SDR ramp in the region of three months, with average tenure not much beyond a year and a half.
Do the arithmetic on a single hire. Three months at a fraction of expected output, in a role that lasts around twenty months, means roughly one eighth of the rep's entire tenure is spent below productivity. Multiply that by a cohort of six and the number stops being an HR statistic.
Bucket three: your best seller, not selling
The standard fix for both problems above is shadowing, which is paid for by removing your strongest rep from selling. It is the most expensive coaching hour available, and it produces observation rather than skill, because watching someone handle a difficult moment does not build the reflex to handle it yourself.
Salesforce's State of Sales research has repeatedly found that reps spend only around 30 percent of their time actually selling. Shadowing hours come out of that 30 percent, not out of the rest.
Why the cost stays invisible
Because there is no invoice. A lost deal is attributed to price, timing, or a competitor, almost never to the fact that the rep was answering that objection for the first time. A slow ramp is attributed to the hire. Neither gets coded as a training failure, so the training budget never comes under pressure and the practice deficit never gets funded.
This is a measurement problem more than a belief problem. Sales leaders generally agree that practice matters. What they cannot see is the specific line where its absence is being paid for, and unmeasured costs do not get fixed.
The counter-argument, taken seriously
There is a real objection to all of this: reps learn from real calls in ways no simulation can reproduce. Reading a room, handling a genuinely unpredictable human, and carrying a complex deal over months are all things you can only develop in the field.
That is correct, and it is not the claim being made here. The distinction is between developing judgement, which needs real calls, and rehearsing a first attempt, which does not. Nobody argues a pilot should not fly real aircraft. The argument is about which manoeuvre they should attempt for the first time with passengers on board.
Practice handles the attempt you can predict. Live calls handle the judgement you cannot rehearse. Confusing the two is what makes training expensive.
What changes when practice is available
Three costs move at once, which is why the return compounds rather than adding up.
- First attempts move off live pipeline. The rep's tenth attempt at the pricing objection is the one the buyer hears, not their first
- Ramp compresses. Repetition becomes available on day two instead of week three, and the waiting weeks disappear
- Manager time redirects. Basic drills stop consuming coaching hours, which then go to deal strategy and judgement
Harvard Business Review, reporting on CSO Insights data, found that organizations with a formally defined sales process saw around 18 percent more revenue growth than those without. A practice layer is how that process stops being a document and becomes a behavior.
How to run the number for your own team
You do not need precision, you need an order of magnitude. Three inputs, ten minutes.
- Count the deals per rep per quarter that ended at the first objection. Ask your reps, they will know.
- Multiply by average deal value, then by rep count, then by four.
- Separately, take your ramp time in months, multiply by monthly quota, and multiply by hires per year.
Add the two figures. In most mid-market teams the result is several multiples of the annual training budget, which is the point: the practice deficit is already being paid for, just not from the line where it would be visible.
Three objections worth answering
Anyone who has run a sales organization will have push-back on this argument. Three versions come up most often, and each deserves a straight answer rather than a dismissal.
"Our reps are experienced, they do not need to rehearse"
Experience covers the conversations a rep has already had many times. It does not cover a new product, a repositioned value story, a new competitor, or a change in pricing, and those happen several times a year in most companies. Every one of them resets a tenured rep to a first attempt.
"Practice is not realistic enough to be useful"
This was a fair objection when practice meant reading a script to a colleague who was being polite. The relevant question now is narrower: is the practice conversation close enough to the real one that the response transfers. For predictable moments like a price objection or a required disclosure, it clearly is. For reading a room across a six month enterprise cycle, it is not, and nobody should claim otherwise.
"We do not have time to add practice"
This one is usually true and misidentifies the trade. Practice does not compete with selling time, it competes with the time currently spent recovering from conversations that went badly and with the shadowing hours that come out of a senior rep's selling week. The question is not whether to spend the time but where.
What the cheapest version looks like
If the argument lands but the budget conversation is months away, there is a version that costs nothing but attention. It will not produce the full effect and it will demonstrate whether the effect is real for your team.
- Pick the single objection your team loses to most, and write the exact sentence buyers use
- Have every rep say their response out loud five times in a Monday meeting, not discuss it
- Score nothing except whether the answer arrived without a pause
- Repeat weekly for a month with the same objection, then measure recovery on live calls
Four weeks of that is enough to see whether repetition changes anything for your team. In most cases it does, which is what makes the case for doing it properly.
The uncomfortable conclusion
Every sales organization already funds practice. Most of them fund it in lost pipeline and slow ramp rather than in a system, and that is the more expensive of the two options by a considerable margin.
The mechanism for changing it is covered in building a practice loop that sticks, the ramp arithmetic in how to reduce sales rep ramp time, and the reason single training events fade in why sales training does not stick.
Frequently asked questions
The visible cost is the training budget. The invisible cost is the pipeline consumed while reps learn on live deals, plus the ramp months during which a new hire produces below target. In most teams the invisible cost is the larger of the two by a wide margin.
Learning on the job is unavoidable and valuable. Attempting something for the first time on a live deal is different, and it is avoidable. The distinction is between developing judgement on real calls, which needs real calls, and rehearsing a first attempt, which does not.
Take your average deal value, multiply by the number of deals per rep per quarter that end at the first objection, and treat that as the annual figure. It is a rough number and it is usually large enough that precision stops mattering.

