Why new hires are still ramping in month four — and how to cut it in half

You hired a senior project manager four months ago. The interviews went well, references checked out, the start was on time. Today she is roughly at sixty percent of expected output. She knows the people. She has read the documentation. She has been in every meeting. Something is still missing, and she would probably struggle to name it precisely. Whatever it is, the company is paying for it.

This is the standard onboarding shape at SMEs, and the standard analysis of it is wrong. People assume the gap is training. It is not. The training was done in the first two weeks. What is missing in month four is context — the accumulated body of project history, client relationships, internal decisions, and recurring patterns that the existing team has in their heads and that the new person is still painstakingly reconstructing one Slack thread at a time.

Training versus context, and why they get confused

When operators talk about "onboarding", they almost always mean two different things that have been collapsed into one word.

Training is teaching someone how to do the job in principle. The tools, the processes, the policies, the standards. This can be designed, measured, completed, and signed off. A well-run training program at an SME takes a couple of weeks.

Context is teaching someone how to do this job at this company, with these clients, after these specific past decisions. The constraints that don't apply at any other company. The history of this account. The reason we abandoned that workflow. The kind of email the head of marketing finds passive-aggressive. The pricing exception we made for one supplier in 2024 that has quietly become standard.

Context cannot be designed and cannot be signed off. It is accreted, one interaction at a time, over the months after training ends. And it is by far the more important of the two.

What month-four really looks like

A new hire at month four typically has full training and partial context. She is technically capable of doing every part of her job. She is not yet reliable at doing it the way the company actually does it.

This produces a specific failure mode. She makes decisions that are technically correct but culturally wrong. She sends emails that are professionally fine but slightly off-tone for the contact. She proposes solutions that have, unbeknownst to her, been tried and abandoned two years ago. Each individual misfire is small. The cumulative effect on the people working with her is "she's good, but she keeps needing to be steered".

The steering is expensive. It consumes the time of more senior people, who do not have it to spare. Worse, the steering is rarely systematic; it happens reactively, one misfire at a time. The new hire is learning, but she is learning at the maximum-friction pace.

Why "shadowing" doesn't scale

The traditional response is shadowing. Pair the new hire with a senior person for the first months. Let her absorb context through observation. This works to a point, mainly because it pulls a senior person off productive work and burns the equivalent of the new hire's salary again in shadowing time.

It also captures context only at the rate of whatever conversations happen to occur while the new person is present. The context she is missing is months and years deep. Shadowing for a quarter exposes her to a quarter's worth of context. The rest she still has to piece together.

The fundamental constraint is that the existing team's context is locked inside the existing team. It is in their heads, their email histories, their private Slack threads, their personal notes. The new hire has no read access to any of it. Every question she might ask requires interrupting someone who has the answer.

The new hire's most expensive resource is everyone else's time

This is the under-appreciated cost. A new hire at month two has, on average, several dozen questions per week that she cannot answer herself. Each question is a small interruption to the person who has the answer. Each interruption is small, but the cumulative weight is real, and it is concentrated on the same handful of senior people who are also the bottleneck for everything else.

Senior people end up dreading new hires not because they don't want to help, but because each new hire adds a low-but-constant interruption tax for six to nine months. This is why the implicit message at most SMEs is "figure it out yourself" — not because anyone is unkind, but because the alternative is unsustainable for the senior team.

What changes when context is readable

The expensive thing about onboarding context is not that it is hard to transfer. It is that it is locked inside people. If the same context were structured and queryable, the new hire could answer the vast majority of her own questions, on her own time, without consuming anyone else's.

Concretely, this looks like the new hire asking, in her own words: "what has been said about the TeamSystem project in the past six months", and getting back a structured answer pulled from the actual emails, chats, and meeting notes. Or: "what are the standing concerns from this client", and getting a pattern across the last year of interactions. Or: "who proposed switching to Adyen and what did we conclude", and getting the original Slack thread plus the follow-up.

The new hire is no longer waiting for context to be transferred to her. She is reading her way into context herself, at her own pace, without burning anyone else's time. The ramp slope changes.

The companies that do this well share two patterns

The first is that they have stopped treating internal communications as private to individuals and started treating them as institutional records. The emails, the chat threads, the call recordings — these are corporate context, not personal correspondence. A new hire joining a team gets read access to that team's communication history, structured and queryable, from day one.

The second is that they have stopped relying on documentation as the primary mechanism for transferring context. Documentation is fine for stable training material. For operational context, the original conversations are what matters, and the question is whether the new person can navigate them efficiently.

The companies that combine these two patterns are routinely reporting onboarding ramp times that are forty to fifty percent faster than the SME average. Not because the training is better. Because the post-training context acquisition stops being the bottleneck.

What month-two actually looks like at sixty percent ramp

The most striking thing, at companies that have done this well, is what month two looks like. The new hire is not waiting on senior people to brief her on the projects. She has briefed herself, by reading across the actual project history. She comes into meetings with context. She asks better questions earlier. The senior team's interruption tax is dramatically lower.

This is, in retrospect, the obvious right shape. The information was always there. What changed is whether the new person could actually get at it without going through a human.

Cutting onboarding time in half is not about training programs or buddy systems. It is about removing the artificial barrier between "what the company knows" and "what the new hire can read". Remove that barrier and the rest takes care of itself.