Ramp is measured in months, and everyone accepts it
Ask ten sales leaders how long a new rep takes to reach quota and you will hear numbers in months, not weeks. Benchmark surveys of SaaS sales organizations have for years put average account executive ramp somewhere in the three to five month range ( The Bridge Group, SaaS AE Metrics). Longer sales cycles, regulated products, and complex catalogs push it further out.
Most of that time is treated as unavoidable. Some of it is. But a meaningful share of it is not about the market at all. It is about how few attempts the rep gets.
Four reasons ramp takes so long
- Onboarding is front-loaded and then forgotten. Week one is a firehose: product, pricing, ICP, CRM, competitors. Very little of it is ever retrieved, and unretrieved material decays fast. Ebbinghaus described the shape of that decay in the 1880s and it has held up ( the forgetting curve). By week three the rep has a certificate and a fog.
- They get almost no real attempts. Do the arithmetic for your own team. In month one, how many genuine discovery conversations does a new rep actually run? For many B2B teams the honest answer is a handful. Nobody becomes competent at anything in five attempts.
- Feedback is rare and late. A manager might sit in on one call in ten, and the debrief often lands days later. By then the rep cannot remember why they said what they said.
- Every mistake costs a real lead. Learning on live buyers is expensive, and the rep knows it. So they play safe, stick to the script, avoid the hard question. Caution protects the pipeline and slows the learning.
What roleplay can take out of live calls
The point of roleplay is not to replace the field. It is to move the repetitive, high-failure parts of the job somewhere failure is free, so the rep arrives at their real calls having already made the obvious mistakes.
Moves well into roleplay
- Discovery sequences and follow-up questions
- Your ten most common objections, in the buyer's actual words
- Price pressure and discount requests
- Competitive comparisons and product edge cases
- Openers, gatekeepers, voicemail, no-show recovery
- Bad news conversations: delays, scope cuts, price changes
Stays in the field
- Real relationships built over months
- Negotiation with people who hold actual authority
- Procurement, legal, and security review
- Account politics and internal champions
- CRM hygiene, handoffs, forecasting discipline
Being clear about the second column matters. A team that expects roleplay to produce a finished seller will be disappointed. A team that uses it to compress the first fifty attempts into the first two weeks usually is not.
A 30-day launch checklist
Before day one, do three things: pick three scenarios from real recorded calls rather than writing generic ones, write down what "ready for live calls" means in observable terms, and name one person who is responsible for the rep's first month.
Week 1. Orientation with retrieval
- Product, ICP, and pricing, but every block ends with the rep explaining it out loud
- First voice roleplay session by day three, deliberately before they feel ready
- Record a baseline score so week four has something to compare against
- Shadow two real calls, with one written observation each
Week 2. Objections
- Fifteen minutes of roleplay a day on the top objections, one at a time
- Same scenario twice: attempt, debrief, immediate second attempt
- Rep reviews their own scores before the manager does
- First supervised live conversation, low-stakes account
Week 3. The whole call
- Full end-to-end conversations, not isolated fragments
- Price pressure and a difficult, interrupting buyer persona
- First solo calls with a same-day debrief
- Manager reviews the trend, not one session
Week 4. Readiness decision
- A certification session scored against the published rubric
- Compare against the week-one baseline on each signal
- Written readiness call: full pipeline, partial, or another two weeks
- Set the ongoing cadence so practice does not stop at day 30
How to tell it is working
Do not judge month one by revenue. Watch leading indicators instead: how many roleplay sessions the rep completed without being chased, whether their scores on structure, questions, objections, and next steps moved between week one and week four, how many days until their first solo call, and what share of their calls end with a real next step.
Then compare cohorts, not benchmarks. The only ramp number that means anything for your team is the one from your last three hires. If the next hire reaches their first booked meeting, or their first closed deal, measurably sooner than that group did, the onboarding change is doing something.
Two ways this usually goes wrong
The first is certifying on knowledge. A quiz proves the rep can recognize the right answer in a quiet room. It says nothing about what they do when a buyer cuts them off. If your onboarding ends with a test, you are measuring the wrong thing.
The second is the opposite: throwing the new rep straight onto live calls to "learn fast." They do learn, slowly, at the cost of leads, and often they learn a defensive habit that takes a year to unwind.
pichi.ai is built to sit between those two. In an AI roleplay a new rep runs a voice conversation with an AI customer built from your real objections, gets a debrief immediately, and goes again. The manager sees who is ready before the rep touches a real account, without listening to a single session.