Draft · Not finished yet
Seeing people while scaling

A Mentorship-Driven Performance Framework

A blueprint for turning growth, feedback and recognition into a system that scales with the company.

“I first got to know this framework when I had the chance to work at @dwsbrazil, and since then I’ve been proposing it to my clients and colleagues. It’s something I’ve lived up close from two perspectives: as an employee, and as someone implementing it across clients and teams.”

@Ivo

Startups scale fast. On purpose.

Funding rounds convert into visible growth. Every one of those moves adds people, teams and complexity.

New funding

Capital secured to build the company’s next stage.

Growing team

New hires across functions; more managers, more layers.

New office & rituals

New space, new processes. The company is being re-founded at scale.

What got a company here won’t scale by itself. Processes for code and product mature early. The process for people usually hasn’t even started.

The gap: performance has no system of record

In most early-stage companies there is no structured performance process. High-value work happens, but nothing systematically tracks, discusses or recognizes it.

What this costs

  • Top performers are left guessing whether their work is seen, and clarity is what keeps great people engaged.
  • Promotions and raises feel arbitrary without shared criteria.
  • Without a regular checkpoint, early signs that someone needs support have nowhere to surface.

Invisible contributions

With no system of record, visibility depends on who happened to be in the room.

Inconsistent decisions

No shared bar for promotion, bonus or feedback.

Silent disengagement

When extra effort and routine effort look the same, motivation is harder to sustain, for anyone.

Retention risk

Replacing one employee costs 50 to 200% of their annual salary (Gallup).

The same discipline as structured hiring

Modern hiring playbooks replace low-signal vibe checks with scorecards, standardized questions and calibrated reference checks. This framework applies the exact same signal-over-noise philosophy after the offer is signed: one continuous system from first interview to annual review.

Scorecard → living document

Every role is hired against a mission, outcomes and characteristics: outcomes, not activities. The living document keeps that scorecard alive after day one: the same outcomes become the growth plan mentor and mentee track all year.

The mentoring promise → Pillar 1

Most offer playbooks promise a mentoring plan and specific growth areas for every A-player. Mentorship from day one is how that promise gets delivered systematically: through onboarding, the first 100 days, and beyond.

Structured interviews → calibrated reviews

The hiring playbook replaces gut feeling with standardized questions and a shared bar. The annual review applies the same discipline internally: one rubric, one panel, comparable ratings.

Hiring finds A-players. This framework is how they keep performing like A-players, and how the first 100 days become the first five years.

The proposal, at a glance

Three connected pieces: lightweight enough for a startup, structured enough to be fair.

1

Mentorship
from day one

Every new hire picks a mentor during onboarding: a guide for growth, separate from their manager.

2

Continuous
1:1 development

Mentor and mentee meet through the year: strengths, gaps, and a living growth plan.

3

Annual calibrated
review

Mentors present each person to a manager panel; a shared 5-level scale drives fair outcomes.

Grow all year  →  evaluate once  →  act on it

Pillar 1: Mentorship from day one

During onboarding

New hires see a curated list of available mentors (senior peers who opted in) and choose one.

Expectations made explicit

The mentor walks through the competencies the team expects: quality bar, collaboration, ownership, growth path.

A guide, not a boss

The mentor’s only job is the mentee’s development. No conflict with delivery pressure or performance judging.

Mentor ≠ Manager

Mentor

A trusted expert crafter: develops skills, advocates for the person, prepares their story for review.

Manager

Owns delivery, priorities and the final evaluation decision with the panel.

Pillar 2: Continuous 1:1 development

A recurring, low-ceremony conversation, monthly or every 6 weeks, owned by mentor and mentee together.

Map strengths & gaps

Honest picture of where the person shines and what blocks the next level.

Living growth plan

2 to 3 concrete goals per cycle, seeded from the role’s hiring scorecard and tied to real work, not abstract courses.

Evidence log (“brag doc”)

Wins, shipped projects and impact recorded as they happen. Nothing lost by December.

Feedback both ways

Course-correct in weeks, not in an annual surprise. Constructive, high-EQ, in both directions. The no-surprises rule starts here.

Pillar 3: The annual performance review

Once a year, a structured calibration session: the industry-standard way to keep ratings fair across teams.

1

Mentor prepares a one-pager

Distilled from the living document built in the 1:1s all year: growth, key deliveries, strengths, open gaps.

2

Panel session

Managers + the person’s mentor meet. The mentor presents the one-pager (~5 min per person).

3

Calibration round

Quick structured discussion; the panel compares against the shared 5-level scale: data over opinions, never the loudest voice in the room.

4

Outcome & follow-up

Each person leaves with a clear rating, the reasoning behind it, and the action attached to it.

Nothing said in the room should surprise the person, because feedback was continuous all year.

The 5-level scale: shared language, concrete actions

Each rating has a clear meaning and a default next step. A rating without a consequence is theater.

What it meansWhat happens next
1
Below the barDoesn’t meet team standards.Formal improvement plan (PIP): 60 to 90 days, explicit goals, weekly check-ins, and an honest exit path if unmet.
2
DevelopingSpecific skills need work, nothing critical.Focused skill plan with the mentor; re-check at mid-year. Goal: reach level 3, fast.
3
Solid, on trackMeets expectations consistently, the healthy core of any team.Recognition that consistency matters; growth plan continues toward the next level.
4
Standout highlightsClearly above average, with impact beyond their own scope.Public recognition, stretch projects, priority in salary review; promotion track begins.
5
Far exceedsPerforming well beyond plan. These people set the culture.Promotion, bonus and/or equity refresh, deliberate and visible.

Grounded in market best practices

Calibration sessions

Rating together, against one rubric, is a strong bias reducer (used across Big Tech).

No-surprises rule

Continuous feedback means year-end confirms what everyone already knows.

Separate growth from judgment

Development talks (mentor) and evaluation (panel) in different rooms, so people stay honest in both.

Evidence over memory

Brag docs and one-pagers beat recency bias: December doesn’t erase March.

Lightweight by design

One page per person, one session per year. Context, not rules. Heavy processes die in startups.

Tied to real outcomes

Ratings connect to comp, promotion and support, so the process earns trust.

What the data says

more likely to be engaged when employees get valuable feedback from the people they work with

Gallup × Workhuman
−30%

voluntary attrition at Adobe within a year of replacing annual reviews with continuous “Check-ins”

Forbes · Stanford GSB case study
41% vs 25%

workers without a mentor vs. with one who considered quitting in the last 3 months, from a survey of 7,940 U.S. workers

CNBC × SurveyMonkey Workplace Happiness Survey, 2019
50 to 200%

of one employee’s annual salary: the cost of replacing them, by Gallup’s own conservative estimate. And 52% of those exits were preventable.

Gallup

The rhythm of a year

Days 1 to 30

Onboarding

New hire picks a mentor; the scorecard’s 30/60/100-day outcomes and competencies made explicit.

All year

1:1 cycles

Monthly development 1:1s; the living document grows; goals updated.

~Month 6

Mid-year check-in

Light temperature check: is the growth plan on track? Any early flags?

Month 11

Review & calibration

One-pagers, panel session, 5-level rating.

Month 12

Outcomes land

Promotions, bonuses, plans, all communicated 1:1 with reasoning.

How to start: a 90-day pilot

Weeks 1 to 4 · Design

Extend the hiring scorecards into a per-level competency rubric; mentor guidelines, one-pager & living-document templates. Volunteer mentors opt in.

Weeks 5 to 12 · Pilot in one team

Pair everyone in the pilot team with a mentor; run two 1:1 cycles; dry-run one calibration session with real one-pagers.

Quarter 2 · Evaluate & expand

Survey pilot participants, tune the rubric, then roll out company-wide before the first real annual cycle.

Cost to try: a few templates, volunteer mentors, and one afternoon of leadership time. No new tools, no new headcount.

Culture, made operational

Most startup culture decks say it plainly: culture is not what’s written, but what’s practiced. A framework like this is that practice: each pillar operationalizes a principle most teams have already committed to on paper.

High expectations, as a gift

High-expectation cultures ask for honesty when something isn’t working, and for addressing it directly. The 5-level scale and its default actions are that honesty, with a fair process around it, protecting the talent density startups win with.

Invest in the talent

Nearly every company claims to invest in talent: leveling people up to level up the impact they can make. Mentorship from day one is that investment, running on a system instead of goodwill.

Less opinions, more data

Every scale-up playbook is explicit: data trumps opinions, and it’s not about the loudest voice in the room. Evidence logs and a calibrated panel apply that same principle to performance: decisions read from a year of data.

Craftsmanship over politics

In craft-driven cultures, authority is earned by craft, and the best managers are trusted expert crafters rather than pure people managers. This framework follows that lead: mentors are senior crafters, and recognition follows outcomes on the record: products, not ladders.

Catch people doing right

“Catch people doing right” asks for quick, specific praise that reinforces what good looks like. The living document does exactly that, all year and on the record. And levels 4 and 5 turn it into deliberate, visible recognition.

Bias for action & plasticity

A 90-day pilot is a startup’s favorite kind of decision: easy to reverse. Do, learn from the doing, and fix forward. And if the framework stops making sense, plasticity says change it.

And it honors “less rules, more context”: a shared rubric isn’t a rule that replaces judgment. It’s the context that lets good judgment make fair, comparable decisions.

References

Every number in this article traces back to one of these sources. Links open in a new tab.

Laszlo Bock, Work Rules! (Twelve, 2015)

Google People Ops: calibration meetings, 5-level ratings, and separating evaluation from development conversations.

Book · ISBN 978-1455554799
Gallup, “This Fixable Problem Costs U.S. Businesses $1 Trillion”

Replacing an employee costs 50 to 200% of their annual salary (a conservative estimate); 52% of voluntary exits were preventable.

gallup.com/workplace/247391/fixable-problem-costs-businesses-trillion.aspx
Harvard Business Review (Jan 2024), Khan, Korn & Williams

“How Calibration Meetings Introduce Bias into Performance Reviews”: calibration helps, but needs structure.

hbr.org/2024/01/how-calibration-meetings-introduce-bias-into-performance-reviews
Forbes (Burkus, 2016) · Stanford GSB case: Adobe “Check-in”

30% drop in voluntary attrition after replacing annual reviews with continuous check-ins.

forbes.com/sites/davidburkus/2016/06/01/how-adobe-scrapped-its-performance-review…
Harvard Business Review (Jun 2018), Demeré & Sedatole

“Why Managers Shouldn’t Have the Final Say in Performance Reviews”: committees correct manager bias.

hbr.org/2018/06/why-managers-shouldnt-have-the-final-say-in-performance-reviews
CNBC × SurveyMonkey Workplace Happiness Survey (Jul 2019)

Survey of 7,940 U.S. workers: 91% of mentored workers are satisfied; 41% without a mentor considered quitting vs 25% with one.

cnbc.com/2019/07/16/nine-in-10-workers-who-have-a-mentor-say-they-are-happy…
Gallup × Workhuman, feedback & recognition research

Valuable feedback → 5× engagement, −57% burnout, −48% likelihood of job hunting.

gallup.com/workplace/651812/organizations-redefine-feedback-including-recognition.aspx
Julia Evans, “Get your work recognized: write a brag document”

The evidence-log practice behind the living document, standard in engineering orgs.

jvns.ca/blog/brag-documents
PT