Time-to-Productivity: How a Company Brain Makes New Hires Productive in Days, Not Months

You hired a strong person. They signed in March. It is now November, and they are finally running at the pace you hired them for. Eight months. Nobody was slacking - the new hire spent those months learning how your company actually works, and your best people spent them answering the same questions over and over instead of doing their own jobs.

This is the quietest, most expensive line item in any growing company, and almost nobody measures it. Oxford Economics put a number on it: a typical role takes 28 weeks to reach optimum productivity, and the lost output during that ramp is the single largest cost of a departing employee, worth 25,181 pounds per person³. Gallup found only 12 percent of employees strongly agree their company does a good job onboarding¹. The knowledge that new hires need is real, but it lives in people’s heads and decays in wikis nobody updates.

This guide is for the operations leader, HR lead, or Geschaeftsfuehrer who is tired of watching every new hire disappear into a months-long fog. The fix is not a better checklist. It is a Company Brain - a living memory of your company’s people-knowledge, processes, and data that survives turnover, answers “how do we do this here” on demand, and lets AI employees take the routine ramp-up work. The team grows in output without growing in headcount.

The 8-Month Problem

Every company measures time-to-hire. Almost none measure what happens after the contract is signed, which is where the real money goes. A new hire is a fixed cost from day one but delivers a fraction of their value for months. That gap - full salary against partial output - is the 8-month problem, and it is getting worse as roles get more complex and tenures get shorter.

• Eight months to full speed - The widely cited cross-industry average for a new hire to reach full productivity is around eight months, and it stretches past a year in complex or senior roles².
• 28 weeks of lost output - Oxford Economics measured 28 weeks to optimum productivity for a typical role and attributed 25,181 pounds of replacement cost to that ramp alone, more than the recruiting cost itself³.
• Onboarding is broadly failing - Only 12 percent of employees strongly agree their organisation does a great job onboarding, and just 29 percent feel fully prepared after it¹.
• Early exits multiply the loss - In Germany, softgarden found roughly one in six new hires leaves within the first three months, and a botched onboarding is estimated to burn around 43,000 euros per head¹¹.
• Ramp is now the dominant cost - Learning platforms and HR analysts agree: the invisible cost of ramp time, not advertising or agency fees, is where turnover really hurts⁹.
• Shorter tenures make it recurring - As tenures shrink, companies pay the ramp cost more often, so a slow ramp compounds across every hire and every backfill¹⁸.

The instinct is to fix this with more onboarding: longer inductions, thicker handbooks, more training modules. That treats the symptom. The disease is that the knowledge a newcomer needs is scattered, tacit, and decaying - and no amount of front-loaded documentation reaches it.

What Time-to-Productivity Actually Is (and Why It Is the KPI That Matters)

Time-to-productivity is the average time from a new hire’s first day until they perform at the full level expected of the role, unaided. It is the one onboarding metric that converts directly into euros, because every day below full output is a day of salary paid for partial return. Josh Bersin argues HR should move on from dated measures like time-to-hire and course completion toward meaningful ones like time-to-productivity⁸.

How it differs from the metrics you already track

How to define the milestones

The metric is only useful if the finish line is concrete. Full productivity is not a feeling; it is a set of role-specific milestones a competent performer hits. Define them per role, then measure the median days each new cohort takes to reach them^16,17.

• Support agent - First solo resolved ticket, then time to handle a full queue at target quality without escalation.
• Software developer - First merged pull request, then the 10th pull request, an established ramp marker used across the industry⁵.
• Sales rep - First qualified opportunity created solo, then first closed deal and time to full quota-carrying capacity.
• Accountant or clerk - First month-end task completed unaided, then a full close cycle owned without review flags.
• Field technician - First job closed solo at first-time-fix target, then full route capacity.

Once the milestones exist, the question becomes mechanical: what is slowing people down between day one and milestone three? The answer is almost always the same, and it is not the work itself.

Why Onboarding Breaks: The Knowledge Lives in Heads and Decays in Wikis

A new hire in a knowledge role rarely struggles with the task in the abstract. An experienced accountant knows accounting. What they do not know is how this company codes a particular cost centre, which approver actually signs off, and the unwritten exception everyone follows but nobody documented. That company-specific, tacit layer is what the ramp is really spent learning - and it is exactly the layer that traditional onboarding cannot deliver.

• The knowledge is tacit - The rules that matter most live in experienced colleagues’ heads as intuition, not in any document. They surface only when a specific situation triggers them.
• Wikis decay from day one - Documentation is written once, then reality moves on. Within months the wiki describes a process that no longer matches how the work is actually done, so new hires learn the wrong thing or stop trusting it.
• The newcomer interrupts to learn - With no living source of truth, the new hire’s only option is to tap a colleague on the shoulder. Every question is an interruption that slows two people, not one.
• Experts are the bottleneck - The people best able to answer are the people with the least spare time. Onboarding quality depends on whoever happens to be free, so it is inconsistent and slow.
• Context is scattered - The answer to a single question can require the CRM, an email thread, a spreadsheet, and a Teams message. A newcomer does not know where to look or that the sources exist.
• It walks out the door - When a veteran leaves, their tacit knowledge leaves with them, so the next new hire ramps against an even thinner base. Turnover makes the problem worse each cycle¹⁴.

The generic AI trap

Handing a new hire a generic chatbot does not solve this either. A model trained on the public internet knows how accounting works in general; it does not know your chart of accounts, your approvers, or your exceptions. As DX CTO Laura Tacho observes, AI tends to surface existing gaps rather than paper over them - a generic assistant grounded in nothing will confidently give a new hire an answer that is right for the world and wrong for your company.

How a Company Brain Onboards the Human

Flip the frame. For years we have onboarded the human into the company by loading them with documents and hoping the tacit knowledge sticks. A Company Brain reverses the direction: the company’s living memory onboards the human. The newcomer draws on it exactly when they need it, and AI employees running on top of it take the routine ramp work off their plate.

What the Company Brain does for a new hire

• Answers “how do we do this here” - The new hire asks in plain language and gets your company’s specific answer, with the source, in seconds instead of hunting or interrupting.
• Carries the tacit context - Because the Brain is fed by real work and corrections, it holds the exceptions and unwritten rules that a wiki never captured.
• Stays current automatically - When the process changes, the change shows up in the work and the feedback, and the Brain updates. The newcomer learns today’s process, not last year’s.
• Cites its sources - Every answer traces back to a document, record, or decision, so the new hire can verify and an expert can correct it once for everyone.
• Protects your experts’ time - The thousand small questions go to the Brain, freeing senior people for the judgement and coaching that actually need a human.
• Lets AI employees run the routine - AI employees grounded in the Brain handle the repeatable ramp tasks - data entry, first-draft work, lookups - so the newcomer moves to real judgement work sooner.

A day-one scenario, before and after

The mechanism is simple and it is the same one that keeps the Brain accurate: the work feeds the memory, and the memory ramps the people. That loop is what a static onboarding stack has never had.

The 2026 Evidence: Days, Not Months

The claim that grounded AI compresses ramp time is not aspirational. The cleanest measurement comes from software engineering, where output is countable, but the broader onboarding data points the same way.

The developer ramp study

DX analysed new-hire ramp across six multinational enterprises using the 10th pull request as an established productivity milestone. The gap between AI users and non-users is large and consistent⁵.

• Nearly half the ramp - Daily AI users reached the 10th pull request in 49 days versus 91 for non-users, a 46 percent reduction⁵.
• Fewer stragglers - Half of new hires without AI still had not reached 10 pull requests after three months, compared with under 20 percent of daily AI users⁵.
• Sustained, not just fast - Daily AI users shipped changes 1.3 times per week versus 0.65 for non-users, double the ongoing deployment frequency⁵.
• The effect scales with usage - The more grounded, everyday the AI use, the faster the ramp, which is exactly what you would expect if the value comes from answering real, in-context questions⁵.

Beyond engineering

• Onboarding completes faster - Organisations deploying AI in onboarding report the process completing around 53 percent faster and new hires productive roughly 40 percent sooner⁷.
• Admin collapses - The same deployments cut administrative onboarding workload by around 75 percent, freeing HR and managers for the human side⁷.
• Structured onboarding already helps - Even without AI, SHRM links structured onboarding to up to 50 percent greater new-hire productivity and 69 percent three-year retention; a Company Brain is structure that maintains itself⁴.
• The market is moving - Onboarding analysts now treat time-to-productivity as the headline 2026 KPI and AI-assisted ramp as a competitive advantage, not an experiment^10,13,20.

The Euro Model: What a Slow Ramp Costs and What Compressing It Recovers

To make the business case, put the ramp on a single spreadsheet. The logic is straightforward: during the ramp you pay full cost for partial output, so the loss is the gap between full productivity and actual productivity, integrated over the ramp period. Here is a defensible model for a mid-sized company. The numbers are illustrative; swap in yours.

The ramp-loss formula

• Fully loaded cost per hire - Take salary plus employer costs. For this model, 70,000 euros per year, or about 5,833 euros per month.
• Ramp length - Assume the industry-typical 8 months to full productivity².
• Average productivity during ramp - A new hire climbing from 0 to 100 percent over 8 months averages roughly 50 percent output across the ramp.
• Lost output per hire - 8 months at 50 percent of 5,833 euros is about 23,300 euros of paid-but-unproductive cost per new hire.
• Hiring volume - A 200-person company with normal turnover and growth hires roughly 30 people a year.
• Annual ramp loss - 30 hires times about 23,300 euros is roughly 700,000 euros a year in ramp cost, most of it invisible on any P&L line.

The costs that hide behind the ramp

• Senior interruption cost - Every question a newcomer asks a senior person is time that senior person is not doing their own high-value work. This is a second, uncounted loss on top of the ramp.
• Early-exit risk - Slow, frustrating ramps drive early quits. In Germany a first-90-days exit is estimated at around 43,000 euros, and you then pay the whole ramp again for the replacement^11,12.
• Quality cost during ramp - Work produced at partial competence carries more errors and rework, a cost that lands on customers and colleagues.
• Opportunity cost - Capacity you are paying for but not getting is capacity you cannot point at the backlog, which caps how fast the whole team can grow output⁹.

The 90-Day Time-to-Productivity Playbook

You do not boil the ocean. You stand up a Company Brain for the one or two roles you hire most, measure the ramp before and after, and let the gains compound. Here is the week-by-week path from decision to a measurably faster ramp.

Phase 1: Baseline and Scope (Weeks 1-4)

1. Week 1: Pick the high-hiring role - Choose the role you onboard most often or where the ramp hurts most. High volume means the Brain pays back fastest and improves fastest.
2. Week 2: Define productivity milestones - Name the three concrete milestones that mark full productivity for that role, and pull the median days recent hires took to reach them. This is your baseline^16,17.
3. Week 3: Map the knowledge sources - Identify where the role’s real knowledge lives: the CRM, the ERP, email threads, wikis, the three people everyone actually asks. Note the tacit rules that appear nowhere.
4. Week 4: Data and governance setup - Confirm access, connect the systems, minimise personal data, run a data protection assessment where needed, and brief the Betriebsrat early. Frame it as removing drudgery from new hires.

Phase 2: Build the Brain (Weeks 5-8)

5. Week 5-6: Ground the Company Brain - Feed it the connected sources and the captured tacit rules. Stand up the AI employees that will run the role’s routine ramp tasks on top of it.
6. Week 7: Expert calibration - Have your best performers in the role ask the Brain real questions and correct it. Every correction makes it sharper and is exactly how it stays current.
7. Week 8: Set the guardrails - Define human review checkpoints, source citations on every answer, and the escalation path for anything the Brain is unsure about.

Phase 3: Ramp a Real Cohort (Weeks 9-12)

9. Week 9: Onboard the next hires on it - The next people to join the role use the Company Brain from day one as their first port of call for “how do we do this here.”
10. Week 10-11: Measure against baseline - Track days to each milestone for the new cohort against the baseline from week 2. Watch senior-interruption volume fall in parallel.
11. Week 12: Report and expand - Put the ramp-time delta and recovered senior hours in front of leadership, then extend the Brain to the next role. The second role is faster because the pattern is proven.

Why This Is Not a Generic Onboarding Tool

The market is full of onboarding software: LMS platforms, checklist apps, HR suites with an “onboarding module.” A Company Brain is a different category. Onboarding tools manage the administrative wrapper around a new hire. A Company Brain closes the knowledge gap that actually drives the ramp.

The eight capabilities that separate a Company Brain

• Living memory - Fed by the actual work, so it reflects how the company operates today, not how someone documented it once.
• Tacit-knowledge capture - Holds the exceptions and unwritten rules that never make it into a handbook.
• In-context answers - Responds to a new hire’s specific question with your company’s specific answer, not a generic one.
• Source citations - Every answer traces back to a record or document, so it is verifiable and correctable.
• Feedback loop - Improves every week as experts correct it and new work flows in.
• AI employees on top - Runs the routine ramp tasks so the newcomer moves to judgement work sooner.
• Turnover resilience - The knowledge stays when people leave, so each new hire ramps against a full base, not a thinning one.
• Cross-system reach - Pulls from CRM, ERP, email, and documents at once, so the newcomer does not need to know where anything lives.

How Superkind Fits

Superkind builds a Company Brain and the AI employees that run on it, tailored to how your company actually works. The approach is process-first, not tool-first: we start from your real workflows and the roles you hire for, not a generic onboarding template you have to bend to.

• Process-first discovery - We map how a role really works and where its knowledge lives before building anything. No assumptions, no off-the-shelf course library.
• Grounded in your systems - The Company Brain connects to your existing CRM, ERP, email, and document stores. Nothing to rip out, nothing new for the team to learn.
• Captures the tacit layer - We work with your best performers to get the unwritten rules and exceptions into the Brain, where every new hire can reach them.
• AI employees for the routine ramp - AI employees take the repeatable first-draft and lookup work, so new hires spend their first weeks on judgement, not busywork.
• Measured against your baseline - We define role milestones and time-to-productivity up front, then prove the ramp reduction against them.
• Survives turnover by design - The Brain is a durable asset. When people leave, the knowledge stays, so the next hire ramps faster than the last.
• Feedback loop built in - Every correction from your experts and every day of work makes the Brain sharper, so ramp times keep falling.
• Security and DSGVO by default - Data stays within your infrastructure via encrypted connections, with access controls and audit trails, and we help scope the Betriebsrat and data protection work.

If you want the wider picture of the pillar this sits on, see our writing on the knowledge half-life, why generic copilots do not know your company, and growing output while headcount stays flat.

Decision Framework: Is a Company Brain Right for Your Ramp?

Not every company should start here today. Use these signals to decide whether compressing time-to-productivity with a Company Brain is your highest-leverage move right now.

Frequently Asked Questions

Henri Jung

Co-founder of Superkind, where he helps SMEs and enterprises deploy custom AI agents that actually fit how their teams work. Henri is passionate about closing the gap between what AI can do and the value it creates in real companies. He believes the Mittelstand has everything it needs to lead in AI - it just needs the right approach.