Project Manager Track
Project Manager
Runs sub-team delivery end-to-end. Owns the plan, surfaces the risks, keeps the client confident.
4levels
·
9competencies
·
Total weight 13.5
·
Hire bar PM2
The 4 PM levels
PM1
Junior
3–5 members. Simple scope, single workstream. Mentor-supervised.
1.8 – 2.4
PM2
Mid
6–12 members. One full sub-team run independently.
2.5 – 3.4
PM3
Senior
12–20 members. Multi-team or small ODC. Candidate for ODC Lead.
3.5 – 4.2
PM4
Principal
20–35 members. Multi-sub-team coordination. ODC Lead path.
4.3 – 5.0
Competency matrix — at a glance
Competency
W
PM1
PM2
PM3
PM4
Delivery Planning & Execution
×2
2
3
3
4
Requirements & Scope Mgmt
×1
1
2
3
4
Risk & Quality Mgmt
×1.5
1
2
3
4
Client & Stakeholder Comm
×2
2
3
4
5
Team Leadership
×1.5
1
3
4
4
Technical Literacy
×1
1
2
3
4
Effort Efficiency (EE)
×1
1
2
3
4
People Management
×1.5
1
2
3
4
Account Health
×2
1
2
3
4
1 Aware
2 Developing
3 Proficient · hire bar
4 Senior
5 Expert
Competency deep-dive
Delivery Planning & Execution
Weight ×2
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"The plan exists; my job is to keep things on it."
The PM knows what planning artifacts look like (sprint plan, burndown, release plan) and can populate them when given inputs. They cannot independently create a plan from scratch. They are learning by doing, not yet contributing.
Observable behaviors
- 1Attends sprint planning, but Tech Lead or senior PM drives the agenda and pace.
- 2Asks the Tech Lead "should this be a 5 or 8?" rather than estimating themselves.
- 3Updates Jira when team members report status — does not chase status proactively.
- 4Sprint board exists but they don't restructure it for the team's needs.
- 5Status reports follow a strict template; off-template situations cause confusion.
- 6Standups feel like a roll-call rather than a working session.
- 7When something slips, the response is to report it up rather than act on it.
- 8Limited use of dashboards beyond default Jira views.
Expected outputs
- Updated Jira tickets reflecting yesterday's status.
- Meeting notes from ceremonies attended.
- A sprint plan populated with team-provided estimates (PM did not estimate).
- Standard weekly status report, template filled in mechanically.
- Auto-generated burndown chart (no commentary).
- Risk log, mostly empty or recording only obvious risks after they appear.
What you would see
The Tech Lead facilitates the sprint planning. The L1 PM updates Jira live during the meeting, captures decisions, occasionally asks clarifying questions, doesn't push back on estimates, and the meeting can comfortably run without them present.
Common traps
- Becomes a glorified scribe — the team treats them as a note-taker.
- Defers all planning judgment to Tech Lead and stays passive too long (over 6 months in a comfortable team).
L2
"I can follow the playbook reliably. When reality deviates, I react."
The PM can run the standard cadence on a small team (3–6 people, mostly known scope) without supervision — they prep meetings, run them, drive Jira, write status reports. They struggle when reality diverges from the standard pattern (scope changes, late ramps, integration issues).
Observable behaviors
- 1Prepares for sprint planning (pre-groomed stories, capacity check).
- 2Runs standup; team participates.
- 3Owns the sprint plan, but estimates still lean on the Tech Lead.
- 4Tracks velocity but treats it as a number to report, not a planning input.
- 5Identifies risks after they appear, not before.
- 6Status reports get written but might miss nuances the client actually cares about.
- 7Burndown is used but not interpreted ("here's the chart").
- 8Slippage triggers a "we'll catch up next sprint" response rather than a re-plan.
- 9Comfortable with Jira day-to-day but doesn't customize the board for the team's needs.
Expected outputs
- Sprint plan, 1–2 sprints out, with team-provided estimates.
- Daily-updated sprint board.
- Weekly status report (template, populated).
- Risk log — known risks listed, often without mitigations.
- Burndown with one-line commentary.
- Velocity tracking spreadsheet (numbers captured, not analyzed).
- Sprint retro notes with action items, many unimplemented.
What you would see
PM opens the meeting, walks through capacity, asks the Tech Lead for top-priority items, team estimates, PM commits sprint scope based on velocity. Meeting runs 90–120 minutes and ends with a populated sprint. The PM doesn't always question whether the right items got prioritized, or whether dependencies are real.
Common traps
- "Hoping" — assumes things will go well rather than planning for what could go wrong.
- Doesn't push back on team estimates ("if Tech Lead says 8, it's 8").
- Misses second-order effects of scope changes (dependencies, testing, deployment).
- Burndown drifts 2–3 days before any intervention.
L3
"I own the plan. The plan is a hypothesis. My job is to make reality match the plan, or replan when reality wins."
The PM can pick up a sub-team, kick it off, build the backlog and release plan, run sprints, and deliver on commitments — without anyone holding their hand. Their estimates are within ±20% on a typical sprint. They spot deviation early and act before it becomes a status-report problem.
Observable behaviors
- 1Drives sprint planning meetings (not just facilitates).
- 2Challenges team estimates with rationale, not just "are you sure?"
- 3Tracks velocity actively and uses it as the planning baseline.
- 4Spots schedule risk 1–2 sprints out and intervenes (replan, descope, or escalate).
- 5Adapts process when the standard isn't working (e.g., adds a mid-sprint sync if scope shifts often).
- 6Pushes back on client scope changes through formal change control.
- 7Status reports adapt to the client's preferences (frequency, format, focus areas).
- 8Knows when to handle locally vs escalate.
- 9Standups are short, focused, and end with actionable next steps.
- 10Sprint retros produce concrete improvements that get implemented over 3+ sprints.
- 11Risk register is living — items added and closed regularly.
Expected outputs
- Sprint plan with rationale (why these items, why this size).
- Release plan covering 2–3 sprints with milestones.
- Active risk register (typically 5–15 items, weekly updated, with owner and mitigation).
- Velocity report with one-line trend analysis.
- Customized client status report (matches client's preferred format/focus).
- Sprint retros with tracked actions sprint-over-sprint.
- Capacity model factoring in vacations, training, and ramp time.
- Dependency map for known cross-team dependencies.
- Sprint kickoff doc for the team (goal, scope, team composition).
What you would see
PM has pre-groomed the backlog with the BA, knows the capacity, and has a draft sprint goal ready. Meeting starts with goal discussion, then team estimates with PM challenging when numbers feel off. Sprint scope is set in 60–90 minutes. PM confirms commitments aloud. Sprint kickoff doc goes to stakeholders within the day.
Common traps
- Process-worship — measures sprint mechanics but loses sight of the business outcome.
- Velocity worship — chases sprint completion at the expense of quality or learning.
- Comfort zone — keeps the team in steady-state too long; doesn't push for stretch.
L4
"I plan multiple horizons simultaneously. I detect failure 3+ sprints out. I am the project's brain, not its calendar."
Beyond running steady-state, the L4 PM plans releases (multi-sprint, with milestones and dependencies), recognizes patterns of failure early, and can recover a project in trouble. They also start to influence how other PMs work.
Observable behaviors
- 1Plans 3–6 sprints out with milestone-based release planning.
- 2Manages dependencies across multiple sub-teams (within the ODC and across ODCs).
- 3Velocity calibration is rigorous — uses historical data with seasonality, not gut.
- 4Recognizes "project in trouble" patterns weeks before they bite (velocity declining, retro items repeating, defect rate creeping).
- 5Knows how to run a project recovery — descope, reshuffle, re-baseline, communicate.
- 6Coaches mid-PMs informally; reviews their plans before client checkpoints.
- 7Adapts agile method to the project (Scrum, Kanban, hybrid) without dogma.
- 8Establishes and revisits team working agreements quarterly.
- 9Drives improvement cycles end-to-end (action items actually close).
- 10Pushes back on senior client stakeholders on scope or timeline with evidence.
- 11Tracks engineering health metrics (deploy frequency, MTTR, defect escape rate), not just velocity.
Expected outputs
- Multi-sprint release plan with milestones and dependency map.
- Velocity calibration model with statistical baseline and outlier handling.
- Project recovery plan (when needed) with re-baselined dates and client communications.
- Engineering health dashboard alongside delivery dashboard.
- Capacity model accounting for skills, ramp, and attrition probability.
- Coaching notes or written guidance for junior PMs.
- Continuous improvement backlog with metrics on the improvements themselves.
- Cross-team dependency tracker with named owners and dates.
- Risk mitigation playbook specific to this client/domain.
What you would see
Sprint planning is fast (45 minutes) because release planning happened separately. The sprint goal ties to a release milestone. The PM has already checked in with each team member individually before the meeting — the meeting is execution, not discovery. Dependencies are flagged with named owners and dates.
Common traps
- "Hero mode" — solves all problems personally, doesn't develop the team or other PMs.
- Over-engineers planning artifacts (too many dashboards, nobody reads them).
- Recovery becomes the default mode rather than prevention.
L5
"My job is no longer to deliver this project — it is to make all projects deliver better. I work on the system, not in it."
The PM designs the standards, templates, and tooling used across the ODC. Other PMs come to them with hard problems. They influence sales (delivery expectations in SOWs) and the ODC's strategic planning.
Observable behaviors
- 1Writes and maintains the ODC planning playbook (versioned, updated quarterly).
- 2Sets estimation standards — reference stories, planning poker conventions, story-point definitions.
- 3Owns or co-owns the Jira configuration, dashboards, and automation.
- 4Brings industry practices in (SAFe, Spotify, Scrum-at-Scale) — selectively adopts, doesn't copy-paste.
- 5Trains and certifies new PMs through structured onboarding.
- 6Audits other PMs' release plans and gives written feedback.
- 7Influences how the ODC sells to clients — sets realistic delivery expectations in proposals.
- 8Identifies systemic planning issues across multiple sub-teams.
- 9Co-creates ODC-level OKRs with the ODC Lead.
Expected outputs
- ODC planning playbook (the canonical reference).
- Estimation reference library — reference stories, calibration tools.
- PM onboarding and training material (multi-week curriculum).
- Process improvement RFCs (proposed changes with rationale and pilot results).
- Industry benchmarking reports — how do we compare on velocity, predictability, defect escape.
- Cross-team planning dashboard (executive view).
- Standard templates library (sprint kickoff, release plan, retro, capacity plan, recovery plan).
What you would see
Rarely in sprint planning meetings — they observe occasionally to coach. When they do attend, they ask second-order questions: "What does our last 6 months of velocity look like? Are we still calibrating against the right baseline? Why is this team's retro producing the same action item we saw in another team three months ago?"
Common traps
- Loses touch with day-to-day execution; becomes too theoretical.
- Over-standardizes — kills team-specific innovation.
- Becomes a bottleneck if too many PMs need their input.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Sprint planning
L1: Attends; Tech Lead drives agenda
L2: Prepares and runs standard cadence
L3: Drives planning with rationale
L4: Plans 3-6 sprints with milestones
L5: Sets ODC planning playbook
Estimation & velocity
L1: Asks Tech Lead for sizing
L2: Tracks velocity as report number
L3: Uses velocity as planning baseline
L4: Rigorous calibration with seasonality
L5: Owns estimation reference library
Risk & deviation response
L1: Reports slippage upward
L2: Spots risks after they appear
L3: Spots risk 1-2 sprints out
L4: Detects failure 3+ sprints out
L5: Identifies systemic patterns across teams
Status & reporting
L1: Strict template, mechanical fill
L2: Standard reports, misses nuance
L3: Adapts reports to client preferences
L4: Engineering health metrics dashboard
L5: Cross-team executive dashboard
Process adaptation
L1: Follows existing process strictly
L2: Runs standard cadence reliably
L3: Adapts process when standard fails
L4: Tailors agile method without dogma
L5: Brings selective industry practices in
Anti-signal · auto-disqualifies the level
Rejects
L1: Becomes glorified scribe / note-taker
L2: 'Hoping' rather than planning for failure
L3: Process-worship; loses business outcome
L4: 'Hero mode' — solves all personally
L5: Loses touch with day-to-day execution
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
User story authoring
L1: Types up stories from BA notes
L2: Writes stories with BA's help
L3: Crisp AC; engineers don't interpret
L4: Stories tied to release outcomes
L5: Sets story standards across ODC
Change control discipline
L1: Tracks raw uncategorized log
L2: Categorizes but inconsistent enforcement
L3: Formal CR with impact assessment
L4: Complex changes; contract-aware
L5: Owns ODC change control playbook
Scope creep response
L1: Doesn't push back
L2: Spots obvious changes only
L3: Pushes back with data
L4: Detects creep 2-3 sprints out
L5: Shapes scope at SOW level
Backlog management
L1: Vague items; loose backlog
L2: Organized but has duplicates
L3: Clean, ranked, no zombies
L4: Release-level scope plan
L5: Cross-team scope dashboard
Client prioritization coaching
L1: Captures client requests verbatim
L2: Clarifies but doesn't think cost
L3: Offers trade-off options ('defer Y')
L4: Coaches must-have vs nice-to-have
L5: Mentors PMs on scope negotiation
Anti-signal · auto-disqualifies the level
Rejects
L1: Stenographer — captures words, not meaning
L2: Accepts client asks without impact check
L3: Procedural overload kills client goodwill
L4: 'Yes' person, or too rigid 'no'
L5: Over-standardizes; loses client nuance
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Risk identification horizon
L1: Reports today's blockers only
L2: Lists common risks; updates monthly
L3: Anticipates project-specific risks
L4: Anticipates strategic/client/regulatory risks
L5: Identifies systemic cross-team risks
Risk register discipline
L1: No register; daily issue log
L2: Basic log, vague 'monitor' mitigations
L3: Active register, owners, due dates
L4: Audits other PMs' registers
L5: Owns ODC risk framework
Quality gates & DoD
L1: Passes QA data through unread
L2: QA owns quality conversation
L3: Defines quality gates per release
L4: Engineering quality metrics (coverage, MTTR)
L5: Sets ODC-wide quality standards
Defect tracking & root cause
L1: No defect interpretation
L2: Defect summary in status report
L3: Tracks escape rate, root cause
L4: Crisis playbook; war-room ready
L5: Cross-team root-cause analysis
Crisis & incident response
L1: Reports issues upward
L2: Escalates when issue surfaces
L3: Holds team to Definition of Done
L4: Leads war-room, drafts client comms
L5: Handles board-level risk events
Anti-signal · auto-disqualifies the level
Rejects
L1: Lives in present — no risk horizon
L2: Log exists but doesn't update
L3: Long risk log, items unactioned
L4: Quality becomes religion; team micromanaged
L5: Theoretical modeling loses ground truth
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Client meeting presence
L1: Listens; speaks only when asked
L2: Runs standup; confident on facts
L3: Runs client meetings solo
L4: Manages C-level steering committees
L5: Trusted advisor across executives
Written communication
L1: Strict template emails
L2: Template populated, sometimes robotic
L3: Matches client format and depth
L4: Executive 1-pagers, decision-focused
L5: Sets ODC client comms standards
Hard conversations
L1: Defers complex talks to senior
L2: Defers escalations to senior PM
L3: Handles delay/defect calls directly
L4: Resolves escalations independently
L5: Coaches PMs through tough calls
Expectation management
L1: Sends status only when asked
L2: Over-promises to please client
L3: Raises concerns before client does
L4: Brokers between conflicting stakeholders
L5: Opens new account opportunities
Reading the room / subtext
L1: Misses sarcasm, indirect dissatisfaction
L2: Sticks to script; doesn't dig
L3: Adjusts tone (formal/casual, tech/biz)
L4: Personal relationships with client leaders
L5: Clients call for unrelated advice
Anti-signal · auto-disqualifies the level
Rejects
L1: Becomes 'ghost PM'; avoids interaction
L2: Over-promises under pressure
L3: Over- or under-communicates bad news
L4: Gets too close to client; advocates against own co.
L5: Over-relied-on; becomes bottleneck
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Morale & engagement
L1: Doesn't notice morale dips
L2: Notices fatigue but doesn't act
L3: Owns morale; acts on dips
L4: Maintains morale under pressure
L5: Sets ODC behavioral norms
1:1s & coaching
L1: No 1:1s; relays direction only
L2: Informal, inconsistent 1:1s
L3: Regular cadence; weekly feedback
L4: Coaches mid-PMs on leadership
L5: Mentors senior PMs on leadership
Conflict resolution
L1: Doesn't intervene in conflict
L2: Mediates surface, not root cause
L3: Structured conflict handling (NVC)
L4: Leads team through major change
L5: Consulted on hard team dynamics
Psychological safety
L1: Standup is mechanical, transactional
L2: Standup efficient but not candid
L3: Retros produce real candor
L4: Builds adaptability into culture
L5: Shapes culture across teams
Developing the team
L1: Task tracker; no development
L2: Avoids hard feedback
L3: Knows each motivator; recognizes wins
L4: Spots and develops emerging leaders
L5: Owns leadership curriculum
Anti-signal · auto-disqualifies the level
Rejects
L1: Becomes a task manager, not leader
L2: Avoids hard feedback to keep comfort
L3: 'Everyone's friend' instead of leader
L4: Hero leader; team dependent on PM
L5: Imposes personal style as universal
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Stack & system knowledge
L1: Knows stack at name level only
L2: Understands high-level architecture
L3: Reads architecture diagrams
L4: Spots architecture trade-offs
L5: Co-leads architecture reviews
Design discussion engagement
L1: Listens; doesn't contribute
L2: Asks clarifying timeline questions
L3: Engages meaningfully in design
L4: Bridges client CTO and our TL
L5: Second opinion on hard decisions
Challenging estimates technically
L1: Trusts every Tech Lead estimate
L2: Discusses feasibility at high level
L3: Challenges with technical rationale
L4: Detects technical risk early
L5: Advises on tech-strategy decisions
NFR & quality concerns
L1: Unaware of NFRs
L2: Names common patterns
L3: Understands NFRs conceptually
L4: Holds tech-debt conversations
L5: Understands modern delivery deeply
Tech-client translation
L1: Defers all tech questions
L2: Reads tech docs unaided
L3: Translates tech-client bidirectionally
L4: Client-facing technical briefings
L5: Influences ODC tech hiring profile
Anti-signal · auto-disqualifies the level
Rejects
L1: Bluffs understanding; engineers detect
L2: Asks too many basic questions
L3: Steps on Tech Lead's design turf
L4: Becomes de-facto architect by default
L5: Forgets they're a PM, not CTO
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Timesheet hygiene
L1: Tracks own/team timesheets accurately
L2: Reports actual vs planned
L3: Sprint EE forecast
L4: Cross-sprint trend with commentary
L5: Sets ODC timesheet standards
EE monitoring cadence
L1: Doesn't monitor during sprint
L2: Monitors weekly; reactive
L3: Mid-sprint forecasting
L4: Sustains 95%+ over releases
L5: Sets ODC EE benchmarks
EE attainment
L1: No EE awareness
L2: Lands 80-90%
L3: Consistently 90%+
L4: Sustains 95%+
L5: Benchmarks all teams
Root-cause diagnosis
L1: Sees EE as backward-looking
L2: Doesn't diagnose causes
L3: Identifies dependency/rework causes
L4: Process changes (better AC, less rework)
L5: Shapes estimation standards
In-sprint intervention
L1: No intervention
L2: Flags overruns after the fact
L3: Replan/reshuffle/descope mid-sprint
L4: Cross-checks quality and morale
L5: Coaches PMs on efficiency techniques
Anti-signal · auto-disqualifies the level
Rejects
L1: Doesn't see EE as their responsibility
L2: Reactive; flags after the fact
L3: EE chasing — pushes team into overtime
L4: EE displaces more important outcomes
L5: EE as headline metric, kills quality
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Hiring & interviewing
L1: Relies on HR/ODC Lead
L2: Joins panels as fit screener
L3: Owns hiring; recommends offers
L4: Attracts senior talent via network
L5: Shapes ODC hiring brand
Onboarding
L1: Doesn't own onboarding
L2: Template welcome doc
L3: Personalized first-week plan
L4: Develops successors from day one
L5: Designs career frameworks
Development & IDPs
L1: No development conversations
L2: Performance review form-filling
L3: Writes IDPs; structured 1:1s
L4: Drives team performance calibration
L5: Key voice in promotions, re-orgs
Flight risk & retention
L1: Hears about issues via HR
L2: Misses early attrition signs
L3: Manages flight risks proactively
L4: Regrettable attrition under 10%
L5: Builds inbound talent pipeline
Performance management
L1: Doesn't address performance
L2: Templated review only
L3: Handles underperformance directly
L4: Mentors mid-PMs on people mgmt
L5: Consulted on contentious promo calls
Anti-signal · auto-disqualifies the level
Rejects
L1: Defers entirely to HR and ODC Lead
L2: Reviews are form-filling exercises
L3: 1:1 as status update; IDP wishlist
L4: Plays favorites; calibration biased
L5: Becomes more HR than PM
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Measurement matrix
3-of-5 majority
Dimension
L1 · Aware
L2 · Familiar
L3 · Proficient
L4 · Advanced
L5 · Expert
Feedback collection
L1: Receives only unsolicited feedback
L2: Asks at milestones
L3: Quarterly CSAT or structured
L4: Sustained CSAT >=8/10 over quarters
L5: Client volunteers testimonials unprompted
Trust & relationships
L1: No proactive relationship work
L2: Reports satisfaction up to ODC Lead
L3: Personal trust with PO and dev lead
L4: Client requests PM by name
L5: Trusted partner to client executives
Expansion / growth opportunities
L1: Doesn't notice expansion signals
L2: Recognizes obvious expansion asks
L3: Surfaces 2-3 opportunities/year
L4: Co-pitches expansions; closes some
L5: Contributes to renewal strategy
Churn / risk sensing
L1: Finds out only via escalation
L2: Doesn't follow up when feedback ignored
L3: Acts on negative feedback same cycle
L4: Senses churn risk early; escalates
L5: Strategic account input to ODC Lead
References & advocacy assets
L1: No reference work
L2: Informal expansion log
L3: Trust cadence with key contacts
L4: Provides references; case studies
L5: Other PMs learn this PM's client style
Anti-signal · auto-disqualifies the level
Rejects
L1: Mistakes 'no complaints' for happy client
L2: Feedback qualitative only; no measurement
L3: Single point of dependence with one contact
L4: Becomes client's PM, not company's PM
L5: Identity wrapped in client relationship
Scoring rule: You are at Level N if you hit Level N in at least 3 of 5 dimensions, AND no dimension is below Level (N−1). Sort levels descending — 3th value is your overall level. If the lowest drops more than 1 below, demote 1 level.
Requirements & Scope Management
Weight ×1
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"Requirements come from the client; my job is to capture them."
The PM captures words but not yet meaning. They type up what the BA discovers and what the client says, without questioning either.
Observable behaviors
- 1Captures meeting notes when the client requests features.
- 2Types up user stories from the BA's discovery sessions.
- 3Doesn't write acceptance criteria independently.
- 4Asks "what does the client want?" rather than "what is the underlying need?"
- 5Tracks change requests but doesn't categorize or assess them.
- 6Doesn't push back on scope creep.
- 7Backlog items are vague (e.g., "add reporting").
Expected outputs
- Story tickets with text copied from notes.
- Raw change request log (uncategorized, no impact assessment).
- Loose backlog with many vague items.
What you would see
In requirements meetings, the PM is the scribe. The Tech Lead and BA do the questioning. The PM only contributes when asked.
Common traps
- Becomes a stenographer — captures words but not meaning.
- Lets the BA own the entire scope conversation.
L2
"I write stories from the BA's discovery. AC are checklists."
Can produce reasonable user stories with the BA's input. Recognizes obvious scope changes but misses the subtle ones (e.g., "a filter" that requires three new endpoints).
Observable behaviors
- 1Writes user stories with clear acceptance criteria (with BA's help).
- 2Tracks the change log with high-level categories.
- 3Recognizes obvious scope changes ("client added a new screen").
- 4Misses subtle scope creep (small asks with big downstream impact).
- 5Backlog is organized but with some duplicates or outdated items.
- 6Doesn't enforce formal change control consistently.
Expected outputs
- User stories with acceptance criteria (mostly clear, some ambiguity).
- Categorized change log (defect / new scope / clarification).
- 1–2 sprints of refined backlog.
- Basic story map.
What you would see
BA leads the grooming session; PM facilitates and captures decisions. When the client floats a new idea, PM asks clarifying questions but doesn't immediately think about cost.
Common traps
- Accepts client requests as commitments without checking impact.
- Treats acceptance criteria as a checklist of features rather than testable behaviors.
L3
"Backlog is product strategy in flight. Every change has a cost and I own that cost conversation."
The PM drives backlog and scope discussions. Enforces formal change control with impact data. Pushes back on scope creep without damaging the client relationship.
Observable behaviors
- 1Runs backlog refinement; drives prioritization conversations with client and BA.
- 2Writes user stories with crisp acceptance criteria that engineers don't need to interpret.
- 3Enforces formal change control — every change request gets assessed for time, cost, and risk impact.
- 4Categorizes changes (defect, new scope, clarification, missed requirement).
- 5Pushes back on scope creep with data ("this requires X more sprints").
- 6Maintains a clean backlog (ranked by value, no zombies).
- 7Uses story-mapping or release-mapping to help the client see priority trade-offs.
Expected outputs
- Refined backlog (1–2 sprints deep, ranked).
- User stories with AC and Definition of Done.
- Change Control Register with impact assessment per item.
- Story map or release roadmap (visual).
- Scope baseline document (what was in the original SOW vs what has changed).
What you would see
PM runs the backlog meeting with ranked priorities. When the client wants to add scope, PM responds with "to add X, we'd need to defer Y or extend by N sprints — which do you prefer?" rather than "yes, we can add that."
Common traps
- Procedural overload — every conversation becomes a change request, client loses goodwill.
- Treats the SOW as gospel rather than a starting baseline.
L4
"Scope is a negotiation. I shape it; I don't just receive it."
Negotiates scope trade-offs at release-planning level. Coaches the client on must-have-vs-nice-to-have thinking. Handles complex changes that ripple through dependencies, NFRs, and contracts.
Observable behaviors
- 1Negotiates scope trade-offs with the client at release-planning level.
- 2Aligns scope to release plan and commercial constraints (margin, headcount).
- 3Detects scope creep 2–3 sprints out through pattern recognition.
- 4Coaches the client on prioritization ("if we do all of this, here's what gets sacrificed").
- 5Handles complex changes that affect dependencies, NFRs, or contract structure.
- 6Builds reusable scope-management artifacts (templates, prioritization frameworks).
Expected outputs
- Release-level scope plan with named trade-offs.
- Scope dashboard (commitments, changes, impact, cumulative drift).
- Negotiation outcomes documented.
- Lessons-learned register for scope-related issues.
What you would see
In a client meeting where new requirements emerge, the PM proposes prioritization options ("if you want X, we drop Y, or we extend by N sprints, or we add headcount with a change order — which fits your goals?") rather than escalating.
Common traps
- "Yes" person under client pressure — accepts scope to keep peace.
- Or, conversely, too rigid — kills the client relationship by always saying no.
L5
"Scope discipline is a system, not a heroic act per project."
Sets ODC-wide change control standards. Influences how proposals and SOWs are scoped. Mentors other PMs on scope negotiation.
Observable behaviors
- 1Shapes scope at SOW level — influences how proposals are written.
- 2Sets ODC-wide change control standard (templates, governance).
- 3Mentors other PMs on scope negotiation and prioritization.
- 4Tracks scope-related metrics across teams (scope-change rate, approval cycle time).
Expected outputs
- ODC change control playbook.
- SOW scope-input templates.
- Cross-team scope dashboard.
What you would see
Reviews and improves how the ODC scopes proposals; consulted on contentious scope battles across accounts.
Common traps
- Over-standardizes — client-specific nuance gets lost in templates.
- Disconnects from current project realities.
Risk & Quality Management
Weight ×1.5
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"Risks are what's blocking us today."
Treats risk as reactive triage. Lives in the present. QA-supplied defect data passes through them without interpretation.
Observable behaviors
- 1Reports issues as they appear in standups.
- 2Follows the QA checklist when given one.
- 3Doesn't anticipate failure modes.
- 4Defect data is captured by QA, not interpreted by the PM.
Expected outputs
- Daily issue log.
- QA-supplied defect report passed through to status.
What you would see
In standup, the PM reports today's blockers. Doesn't proactively ask "what could go wrong this sprint?"
Common traps
- Lives in the present — no risk horizon.
- Treats risk as reactive triage.
L2
"I keep a risk log."
Maintains a basic risk log with common risks (timeline, dependency, key person). Updates infrequently. Mitigations are wishful ("monitor").
Observable behaviors
- 1Identifies common risks (timeline slippage, dependency, key-person dependency).
- 2Runs a basic risk log (10–15 items).
- 3Updates monthly or less often.
- 4Quality conversation comes from QA, not from the PM.
- 5Mitigations are vague ("monitor", "escalate if needed").
Expected outputs
- Risk log with status field.
- Defect summary in weekly status report.
What you would see
PM adds risks after they appear in standup; risks rarely have concrete mitigations or owners.
Common traps
- Log exists but doesn't update.
- Mitigations are wishful ("monitor") rather than actionable.
L3
"Risk is an active discipline. Quality is a contract with the client."
Maintains a proactive risk register with named mitigations and owners. Defines quality gates per release. Tracks defect escape rate and acts on trends.
Observable behaviors
- 1Maintains a proactive risk register with named mitigations and owners.
- 2Defines quality gates per release (entry and exit criteria).
- 3Reviews the risk register weekly with the Tech Lead.
- 4Tracks defect escape rate and root-cause categories.
- 5Anticipates risks specific to this project (integration, NFR, third-party SLA).
- 6Holds the team accountable to Definition of Done.
Expected outputs
- Active risk register (\~15–20 items, weekly updated, with mitigation, owner, and due date).
- Quality gates documented per release.
- Defect dashboard (escape rate, root cause, trend).
- Release readiness checklist.
What you would see
In sprint planning, the PM raises 1–2 risks for this sprint. In standup, the PM asks "anything that could derail this?" Quality is part of every status report, not just a footnote.
Common traps
- Risk log becomes long but unactioned (10+ items with no movement).
- Quality treated as "did QA sign off?" rather than "what's our actual defect trend?"
L4
"I anticipate strategic risks. Quality culture is my responsibility."
Anticipates strategic risks (client politics, regulatory shifts, technology changes). Leads crisis recovery when defects hit production. Drives a quality-first culture in the team.
Observable behaviors
- 1Anticipates strategic risks (client politics, regulatory shifts, technology changes).
- 2Leads crisis recovery — declares incident, runs war room, drafts client comms.
- 3Drives quality-first culture in the team (TDD adoption, code review depth, NFR sign-off).
- 4Audits other PMs' risk registers and gives feedback.
- 5Establishes engineering quality metrics (cyclomatic complexity, test coverage, MTTR).
Expected outputs
- Strategic risk briefings to the ODC Lead.
- Crisis playbook (steps to follow when production breaks).
- Engineering quality dashboard broader than defect rate.
- Quality improvement RFCs.
What you would see
When a critical defect hits production, the PM has a war-room running within 2 hours with named owners, blast-radius assessment, and client comms drafted before the client asks.
Common traps
- Quality becomes religion; team feels micromanaged.
- Risk register becomes a "scary list" without action.
L5
"Risk and quality are systems we engineer, not events we react to."
Builds the ODC's risk framework — categories, escalation paths, governance. Defines ODC-wide quality standards. Handles client-board-level risks.
Observable behaviors
- 1Builds the ODC risk framework — categories, escalation paths, governance.
- 2Defines ODC-wide quality standards (entry/exit criteria, code review norms, NFR baselines).
- 3Handles client-board-level risks (contract risk, security risk, reputational).
- 4Sets ODC quality metrics with the Head of Delivery.
Expected outputs
- ODC risk framework documentation.
- ODC quality standards.
- Quarterly risk and quality report to the Head of Delivery.
- Cross-team root-cause analysis.
What you would see
Reviews systemic risks across teams; intervenes when patterns appear across multiple sub-teams.
Common traps
- Theoretical risk modeling that loses ground truth.
- Standards become so heavy they slow teams down.
Client & Stakeholder Communication
Weight ×2
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"I send status when asked; I escalate when stuck."
Limited client interaction. Defers complex conversations to senior PM. Misses subtext.
Observable behaviors
- 1Participates in client calls but rarely speaks unprompted.
- 2Writes basic status emails using strict templates.
- 3Defers complex conversations to senior PM or ODC Lead.
- 4Misses subtext in client messages (sarcasm, indirect dissatisfaction, urgency cues).
- 5Doesn't proactively ping client.
Expected outputs
- Template-based status emails.
- Factual meeting notes.
What you would see
In client calls, the PM listens. If asked a question, they give a short factual answer. They don't initiate conversation.
Common traps
- Avoids client interaction; becomes a "ghost PM".
- Takes everything literally — misses the real concern.
L2
"I run the standup with the client. I write weekly status."
Comfortable in routine client interactions. Status reports are complete. Struggles with non-routine situations.
Observable behaviors
- 1Runs client standup; speaks confidently on factual topics.
- 2Writes weekly status report from template, well-populated.
- 3Handles routine clarifications.
- 4Defers escalations to senior PM.
- 5Can lose nuance in writing (sounds robotic or overly hedged).
- 6Sometimes over-promises to please the client.
Expected outputs
- Weekly status reports (template, complete).
- Meeting agendas for client calls.
- Routine email responses.
What you would see
Client standup runs smoothly; PM keeps to the agenda. "Any blockers?" is the most challenging question they handle confidently.
Common traps
- Sticks to script; doesn't dig when client says something subtle.
- Over-promises under pressure to keep things smooth.
L3
"I own the client relationship at the project level. I handle hard conversations."
Runs client meetings solo. Handles delays and defect conversations. Manages expectations proactively.
Observable behaviors
- 1Runs client meetings independently (planning, status, escalations).
- 2Handles delay and defect conversations directly with the client.
- 3Manages client expectations proactively (raises concerns before the client does).
- 4Writes status reports that match the client's preferred format and depth.
- 5Reads the room and adjusts tone (formal vs casual, technical vs business).
- 6Knows when to escalate to senior PM or ODC Lead.
- 7Builds personal relationships beyond email (informal check-ins, follow-ups on personal mentions).
Expected outputs
- Customized client status reports.
- Escalation memos with proposed options.
- Meeting agendas and structured follow-ups.
- Client communications log.
What you would see
In a tough call (e.g., announcing a 2-week delay), the PM frames the issue, proposes options, takes the heat without deflecting. Client trusts them.
Common traps
- Over-communicates — drowns client in updates.
- Or under-communicates when uncomfortable about bad news.
L4
"I'm a senior peer to client leaders. I solve things at their level."
Manages C-level stakeholders. Brokers between conflicting client voices. Writes executive-level summaries. Negotiates without escalating.
Observable behaviors
- 1Manages C-level client stakeholders (VP and Director level).
- 2Resolves escalations independently.
- 3Brokers between conflicting client stakeholders (Product vs Engineering, etc.).
- 4Writes executive-level summaries (1-page, decision-focused).
- 5Has personal relationships with key client leaders.
- 6Negotiates dates, scope, or terms without escalating to ODC Lead.
Expected outputs
- Executive briefings (1-pagers).
- Documented escalation resolutions.
- Senior stakeholder map with relationship status.
What you would see
In a steering committee, the PM speaks confidently to the client CTO, proposes options, and the CTO trusts the PM's framing of the trade-offs.
Common traps
- Gets too close to the client — becomes their advocate against own company.
- Or becomes too "company person" — client feels unheard.
L5
"I'm a trusted advisor. Clients ask me strategic questions, not project status."
Trusted advisor to client executives. Opens new account opportunities. Coaches other PMs.
Observable behaviors
- 1Trusted advisor to client executives across multiple stakeholder levels.
- 2Opens new account opportunities through credibility.
- 3Coaches other PMs on client communication.
- 4Influences the ODC's communication standards (templates, escalation paths).
Expected outputs
- Strategic client briefings.
- PM client-communication coaching material.
- Client comms standards for the ODC.
What you would see
Clients call this PM directly even when they're no longer on the account — to ask for advice on unrelated topics.
Common traps
- Over-relied-on; becomes a bottleneck if every client wants to talk to them.
- Identity wrapped in being the trusted person — struggles to step back.
Team Leadership (Daily Dynamics)
Weight ×1.5
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"I track tasks and relay direction."
Reminds team of deadlines, captures action items, relays direction from above. Doesn't intervene in team dynamics.
Observable behaviors
- 1Reminds the team of deadlines.
- 2Captures action items.
- 3Relays direction from above.
- 4Doesn't intervene in team conflict.
- 5Doesn't notice morale dips.
Expected outputs
- Task tracker.
- Meeting follow-ups.
What you would see
Standup is mechanical. Team participates because they have to, not because the PM is leading.
Common traps
- Becomes a task manager, not a leader.
- Misses early signs of disengagement.
L2
"I run good standups and surface blockers."
Effective at meeting mechanics. Recognizes problems but doesn't always act on root causes.
Observable behaviors
- 1Runs effective standups (focused, time-boxed).
- 2Surfaces team blockers and escalates.
- 3Handles minor conflicts ad-hoc when they boil over.
- 4Recognizes when the team is tired but doesn't act on it.
- 5Has informal 1:1s but inconsistently.
Expected outputs
- Standup notes with blockers and owners.
- Ad-hoc 1:1 notes (sparse).
What you would see
Standup is efficient. When a real conflict happens, the PM looks uncomfortable and tries to mediate but doesn't address root causes.
Common traps
- Mediates surface conflict but doesn't address root cause.
- Avoids hard feedback to keep the team comfortable.
L3
"I own team morale day-to-day. I coach individuals. I create psychological safety."
Notices and acts on morale dips. Coaches juniors. Handles conflict with structure. Holds regular 1:1s.
Observable behaviors
- 1Owns team morale day-to-day; notices and acts on dips.
- 2Coaches juniors actively (gives constructive feedback weekly).
- 3Builds psychological safety — team raises hard things in retros.
- 4Handles conflict directly with structure (mediated 1:1, NVC framing).
- 5Holds regular 1:1s (weekly or bi-weekly) with each team member.
- 6Celebrates wins and recognizes effort.
- 7Knows each team member's motivators.
- 8Protects the team from external chaos (scope thrash, exec drive-bys).
Expected outputs
- 1:1 notes per team member on a regular cadence.
- Recognition log.
- Conflict resolution notes.
- Team morale check-ins (formal pulse or informal).
What you would see
Retro produces real candor — the team raises hard things, the PM holds the space, and concrete actions emerge that actually get implemented.
Common traps
- Avoids hard feedback to be liked.
- Becomes "everyone's friend" instead of leader.
L4
"I build culture. The team consistently outperforms peers."
Builds team culture, norms, and rituals. Leads through change. Spots and develops emerging leaders. Maintains morale under pressure.
Observable behaviors
- 1Builds high-performing team culture (team identity, norms, rituals).
- 2Team consistently outperforms peers on velocity, quality, satisfaction.
- 3Leads through change (re-orgs, technology shifts, client changes).
- 4Coaches mid-PMs on leadership.
- 5Spots and develops emerging leaders within the team.
- 6Maintains morale under pressure (during recovery, scope changes).
Expected outputs
- Team charter / working agreements.
- Leadership coaching notes for junior PMs.
- Team-of-team dashboards (if managing multiple).
What you would see
When a major change hits (client changes scope, key engineer leaves), the team adapts quickly and stays motivated — because the PM has built that adaptability into the culture.
Common traps
- Hero leader — team is dependent on PM, PM doesn't develop replacements.
- Culture becomes inward-looking; team feels exclusive.
L5
"I shape culture across multiple teams."
Sets behavioral standards for the ODC. Mentors senior PMs on leadership. Co-creates people strategy with ODC Lead and HR.
Observable behaviors
- 1Shapes culture across multiple teams in the ODC.
- 2Sets behavioral standards for the ODC.
- 3Mentors other senior PMs on leadership.
- 4Co-creates ODC people-strategy with the ODC Lead and HR.
Expected outputs
- ODC behavioral norms document.
- Leadership development curriculum.
What you would see
Other senior PMs come to this person when they have a hard team-dynamics situation.
Common traps
- Becomes too theoretical; loses touch with team realities.
- Imposes their personal style as the universal standard.
Technical Literacy
Weight ×1
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"Tech is the team's job. I track tasks."
Knows the stack at name level. Can describe what the system does at a high level. Can't follow most technical discussions.
Observable behaviors
- 1Knows the project tech stack at name level (React, Java, AWS).
- 2Can describe what the system does at high level.
- 3Can't follow most technical discussions in detail.
- 4Trusts every estimate from the Tech Lead without question.
Expected outputs
- High-level project description.
What you would see
In design reviews, the PM listens but doesn't contribute. When asked a technical question by the client, they defer to the Tech Lead.
Common traps
- Disengaged in technical discussions; team loses respect.
- Bluffs technical understanding — engineers detect it quickly.
L2
"I can discuss feasibility at high level."
Understands high-level architecture. Can discuss feasibility with the Tech Lead. Reads tech docs without translation.
Observable behaviors
- 1Understands high-level architecture (components and interactions).
- 2Discusses feasibility with the Tech Lead ("can we add this feature?").
- 3Reads tech documentation without translation help.
- 4Can name common patterns (REST, microservices, event-driven).
Expected outputs
- Project architecture summary (1-page).
What you would see
In design review, the PM asks clarifying questions about timeline impact of decisions but doesn't yet challenge the technical reasoning.
Common traps
- Bluffs deeper understanding than they have.
- Asks too many basic questions in technical meetings.
L3
"I challenge estimates with reasons. I think about NFRs."
Challenges estimates with technical rationale. Engages meaningfully in design discussions. Understands NFRs conceptually.
Observable behaviors
- 1Challenges estimates with technical rationale ("that's 8 because of integration, right?").
- 2Engages meaningfully in design discussions.
- 3Understands NFRs (performance, security, scalability) at a conceptual level.
- 4Can read a simple architecture diagram.
- 5Knows when to involve the Security or Cloud practice.
- 6Translates tech-to-client and client-to-tech bidirectionally.
Expected outputs
- Architecture decision log (co-owned with Tech Lead).
- NFR checklist for releases.
- Technical risk callouts in status reports.
What you would see
In a tech review, the PM asks "what's our test strategy for the API layer? What's our fallback if the third-party SLA dips?" Engineers respect their input.
Common traps
- Plays technical detective when not needed; slows decisions.
- Tries to make design decisions (steps on Tech Lead's turf).
L4
"I bridge the client's tech leaders and our TL on trade-offs."
Detects technical risk early. Bridges client tech leaders and ODC Technical Leader on trade-offs. Holds informed conversations about tech debt.
Observable behaviors
- 1Detects technical risk early (scaling issues, security gaps, scaling unknowns).
- 2Bridges client tech leaders and the ODC Technical Leader on trade-offs.
- 3Holds informed conversations about tech debt and refactoring.
- 4Spots architecture/quality trade-offs (e.g., monolith vs microservices for this project).
Expected outputs
- Tech-debt register.
- Client-facing technical briefings.
What you would see
In a 3-way conversation between client CTO and our Tech Lead, the PM mediates and proposes options that respect both sides' concerns.
Common traps
- Tries to make architecture decisions (steps on Tech Lead's turf).
- Becomes the de-facto architect by default — engineers stop owning architecture.
L5
"I could pass as a hands-off tech lead."
Advises on tech-strategy decisions. Understands modern delivery deeply. Co-leads architecture reviews.
Observable behaviors
- 1Advises on tech-strategy decisions.
- 2Understands modern delivery deeply (cloud, AI, DevOps).
- 3Co-leads architecture reviews with the Tech Lead.
- 4Influences tech-hiring profile for the ODC.
Expected outputs
- Tech-strategy input to ODC plans.
- Architecture review feedback.
What you would see
Acts as a second opinion to the Tech Lead on hard architecture decisions; consulted on tech hires.
Common traps
- Forgets they're a PM, not a CTO.
- Engineers feel second-guessed.
Effort Efficiency (EE)
Weight ×1
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"Timesheets are HR's thing."
Tracks own and team timesheets accurately. Knows what EE means. Doesn't yet monitor it.
Observable behaviors
- 1Tracks own and team timesheets accurately.
- 2Knows what EE means and how it is calculated.
- 3Doesn't monitor EE actively during the sprint.
- 4Sees EE as a backward-looking metric.
Expected outputs
- Accurate timesheets for self and team.
What you would see
EE doesn't appear in their status reports or sprint reviews unless someone asks for it.
Common traps
- Doesn't see EE as their responsibility.
- Pushes timesheet discipline to HR.
L2
"I report EE weekly; I flag overruns."
Monitors EE weekly and reports actual vs planned. Raises overruns after they happen. EE typically 80–90%.
Observable behaviors
- 1Monitors EE weekly.
- 2Reports actual vs planned in status.
- 3Raises overruns after they happen.
- 4EE typically lands at 80–90%.
- 5Doesn't yet diagnose root causes.
Expected outputs
- Weekly EE report.
What you would see
PM flags an overrun at the end of the sprint and asks the team to be more efficient next sprint, without changing anything structural.
Common traps
- Reactive — flags after the fact rather than acting in-sprint.
- EE becomes a finger-pointing exercise.
L3
"I manage EE in-sprint. 90%+ consistently."
Proactively manages EE during the sprint. Anticipates overruns and acts (replan, reshuffle, descope). Consistently 90%+.
Observable behaviors
- 1Proactively manages EE during sprint (replan, reshuffle, descope).
- 2Anticipates man-month overruns and acts mid-sprint.
- 3Consistently lands at 90%+ EE.
- 4Identifies root causes when EE dips (estimation accuracy, dependency, rework).
Expected outputs
- Sprint EE forecast.
- Mid-sprint replan when needed.
- Root-cause notes when EE dips.
What you would see
At sprint midpoint, the PM has already noticed that one workstream is trending toward an overrun and has redirected a person or descoped an item.
Common traps
- Optimizes for EE at the cost of quality (rushes the team, skips testing).
- EE chasing — pushes team into overtime to hit the number.
L4
"I drive EE improvement through process. 95%+."
Drives EE improvement through process changes (better estimation, fewer dependencies, less rework). Sustains 95%+. Cross-checks against quality and morale.
Observable behaviors
- 1Drives EE improvement through process (better estimation, fewer dependencies, less rework).
- 2Sustains 95%+ over multiple releases.
- 3Cross-checks EE against quality and morale — doesn't sacrifice either.
- 4Mentors juniors on EE management.
Expected outputs
- EE improvement initiatives (with measured impact).
- Cross-sprint EE trend with commentary.
What you would see
Quarterly review shows EE trending up by structural improvements (e.g., reduced rework via better AC), not by squeezing the team harder.
Common traps
- Over-optimizes; team feels squeezed.
- EE becomes the headline metric, displacing more important outcomes.
L5
"I set ODC EE benchmarks."
Sets EE benchmarks for the ODC. Coaches other PMs on efficiency techniques. Shapes estimation standards.
Observable behaviors
- 1Sets EE benchmarks for the ODC.
- 2Coaches other PMs on efficiency techniques.
- 3Shapes estimation standards across teams.
Expected outputs
- ODC EE standards document.
- Efficiency coaching material.
What you would see
Consulted when a project's EE is consistently below the ODC benchmark.
Common traps
- Pushes EE as the headline metric, displacing quality and morale.
People Management (Hire · Develop · Retain)
Weight ×1.5
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"HR handles people."
Knows team headcount. Submits HR paperwork. Relies on ODC Lead and HR for hiring decisions.
Observable behaviors
- 1Knows team headcount.
- 2Submits HR paperwork.
- 3Relies on ODC Lead and HR for hiring.
- 4Doesn't yet own onboarding.
Expected outputs
- Completed HR paperwork.
What you would see
When a team member is unhappy, the PM finds out from HR or the ODC Lead, not directly.
Common traps
- Defers entirely to HR and ODC Lead.
- Misses early signs of attrition risk.
L2
"I join interviews and do onboarding."
Joins interview panels (fit screens). Runs basic onboarding. Runs scheduled performance reviews from template.
Observable behaviors
- 1Joins interview panels (fit screens, not technical).
- 2Does basic onboarding (welcome doc, first-week plan).
- 3Runs scheduled performance reviews from template.
- 4Doesn't yet write Individual Development Plans.
Expected outputs
- Onboarding plans.
- Performance review forms (completed).
What you would see
Performance review is form-filling — boxes checked, ratings assigned, but no deep conversation.
Common traps
- Treats reviews as form-filling exercises.
- Onboarding is template-driven, not personalized.
L3
"I own my sub-team's people. Hire well. Develop them. Notice flight risks early."
Owns sub-team hiring. Runs structured monthly 1:1s. Writes IDPs. Identifies flight risks proactively.
Observable behaviors
- 1Owns sub-team hiring (defines profile, interviews, recommends offers).
- 2Runs structured monthly 1:1s with every team member.
- 3Writes Individual Development Plans (IDPs) for each.
- 4Identifies and manages flight risks proactively.
- 5Recognizes good performance through a structured recognition log.
- 6Handles underperformance directly, not by avoidance.
Expected outputs
- Hiring scorecards.
- Monthly 1:1 notes per team member.
- Individual Development Plans.
- Flight-risk log.
- Recognition log.
What you would see
A team member feels demotivated — the PM has already noticed (from the weekly 1:1 cadence) and had a structured conversation before the issue becomes attrition.
Common traps
- Treats 1:1 as status update; doesn't dig.
- IDPs become wishlists without follow-through.
L4
"I drive calibration. I develop successors. I attract senior talent."
Drives team performance calibration. Develops succession. Attracts senior talent via network. Keeps regrettable attrition under 10%.
Observable behaviors
- 1Drives team performance calibration with peer PMs.
- 2Develops succession — the next PM or Tech Lead from within.
- 3Attracts senior talent through personal network.
- 4Keeps regrettable attrition under 10%.
- 5Mentors mid-PMs on people management.
Expected outputs
- Performance calibration outcomes.
- Succession plan.
- Attrition and flight-risk dashboard.
- Talent pipeline (personal network).
What you would see
When a key engineer leaves, there is already a successor in waiting because the PM had been developing that person for 12 months.
Common traps
- Plays favorites; calibration becomes biased.
- Develops successors but doesn't actually delegate to them.
L5
"I shape ODC talent strategy."
Shapes ODC talent strategy. Designs career frameworks. Builds talent pipeline through community and branding.
Observable behaviors
- 1Shapes ODC talent strategy with the Head of Delivery.
- 2Designs career frameworks (this kind of document).
- 3Builds talent pipeline through community involvement and personal branding.
- 4Key voice in offers, promotions, and re-orgs.
Expected outputs
- Career framework documents.
- Hiring brand materials.
- ODC people-strategy inputs.
What you would see
Speaks at industry events, attracts inbound applicants, and is the person ODC Lead consults on a contentious promotion call.
Common traps
- Becomes more HR than PM; loses delivery edge.
Account Health (Project Level)
Weight ×2
L1
1
L2
2
L3 · hire bar
3
L4
4
L5
5
L1
"Feedback comes when the client gives it."
Receives feedback when given. Doesn't actively ask. Knows what CSAT and NPS are but doesn't track them.
Observable behaviors
- 1Receives client feedback when it comes.
- 2Does not actively ask for feedback.
- 3Knows what CSAT and NPS are but does not track them.
Expected outputs
- Records of unsolicited feedback.
What you would see
The PM doesn't know whether the client is happy or unhappy until something escalates.
Common traps
- Treats feedback as the client's job.
- Mistakes "no complaints" for "happy client".
L2
"I ask for feedback at milestones."
Asks for feedback at release milestones. Reports satisfaction up to ODC Lead. Recognizes obvious expansion signals.
Observable behaviors
- 1Asks for feedback at milestones.
- 2Reports satisfaction to the ODC Lead.
- 3Recognizes obvious expansion signals ("we should also add reporting...").
- 4Doesn't yet measure systematically.
Expected outputs
- Milestone feedback summaries.
- Informal expansion-signals log.
What you would see
After a release, PM sends a feedback email to the client. Response or no response, the PM moves on.
Common traps
- Feedback is qualitative only — no measurement.
- Doesn't follow up when feedback is ignored.
L3
"I measure satisfaction. I spot growth. I build trust."
Measures client satisfaction at least quarterly. Builds personal trust with client product owner and dev lead. Surfaces 2–3 project-level expansion opportunities per year.
Observable behaviors
- 1Measures client satisfaction at least quarterly (CSAT or structured feedback).
- 2Builds personal trust with the client Product Owner / dev lead.
- 3Surfaces 2–3 project-level expansion opportunities per year to ODC Lead.
- 4Acts on negative feedback within the same cycle.
Expected outputs
- Quarterly CSAT or structured feedback survey results.
- Expansion-opportunity log.
- Trust-building cadence (informal check-ins with key client contacts).
What you would see
Client product owner mentions wanting to add reporting features in a status call. PM logs it, follows up with ODC Lead within the day, and the ODC Lead pursues a change order.
Common traps
- Treats CSAT as a metric to chase rather than a signal to act on.
- Builds trust with one client contact only (single point of dependence).
L4
"CSAT ≥ 8/10. Client requests me by name. I co-pitch growth."
Sustains CSAT ≥ 8/10. Client requests this PM specifically. Co-pitches expansions with ODC Lead. Senses churn risk early.
Observable behaviors
- 1Sustains CSAT ≥ 8/10 (or NPS ≥ 50) over multiple quarters.
- 2Client requests this PM specifically for new work or new phases.
- 3Co-pitches expansions with the ODC Lead and closes some independently.
- 4Provides references on request.
- 5Senses churn risk early — escalates before the issue becomes contract-level.
Expected outputs
- Quarterly CSAT trend report.
- Expansion pipeline shared with ODC Lead.
- Reference and case-study materials.
What you would see
When a new SOW is being pitched, the client asks "is \[PM Name\] going to be on it?" before signing.
Common traps
- Becomes the client's PM more than the company's PM (advocates for client against own org).
- Over-invests in one account — bench versatility drops.
L5
"Client treats me as a trusted partner."
Trusted partner to client executives. Contributes to renewal and expansion strategy. Volunteered case studies and testimonials.
Observable behaviors
- 1Trusted partner to client executives — consulted on broader topics.
- 2Contributes meaningfully to renewal and expansion strategy.
- 3Client volunteers case studies and testimonials unprompted.
- 4Other PMs in the ODC learn from this PM's client style.
Expected outputs
- Volunteered case studies and testimonials.
- Mentoring artifacts on client relationship.
- Strategic account input to ODC Lead.
What you would see
Client invites this PM to internal client strategy sessions; references them in public talks or case studies without being asked.
Common traps
- Over-investment in a single account locks the PM in place.
- Identity gets wrapped in the client relationship.
How to advance — promotion criteria
PM1PM2
Typical 18–24 months
What to demonstrate
- L3 in Delivery, Client Comm, Team Leadership
- L2+ in all other competencies including Account Health
- Run a 6+ member sub-team for 2 consecutive sprints
- Sustain EE ≥ 90% for 3 consecutive sprints
Portfolio (CORE)
- 3 recent weekly status reports
- 1 sprint plan with rationale
- Mentor's written endorsement (1 page)
- + pick 2 supporting items
Verification path
1Self-assessment using the worksheet
2Submit portfolio + manager calibration
3Annual review passes with no "below expectation"
PM2PM3
Typical 24–36 months
What to demonstrate
- L3 in all 9 competencies
- L4 in Client Comm AND Team Leadership
- L3+ in People Management
- EE ≥ 92% for 6 consecutive sprints
- Recovered or stabilized ≥ 1 troubled sub-team
Portfolio (CORE)
- Multi-sprint release plan with milestones
- 1 client reference letter or testimonial
- 1 sprint recovery example (caught and resolved)
- Self-assessment + manager calibration
- + pick 2 of 5 supporting items
Verification path
1Self-assess + manager calibration
2Submit portfolio
3ODC Lead + Head of D3 calibration review
PM3PM4
Typical 24–36 months
What to demonstrate
- L4 in all 9 competencies (including Account Health)
- L5 in Client Comm OR Team Leadership
- Lead multi-sub-team or small ODC for 12+ months
- Develop a successor (next PM3) from within team
- Regrettable attrition < 10% for 12 months
Portfolio (CORE)
- 1 full project recovery case write-up
- 1 successor development plan with PM3-readiness evidence
- 2 client reference letters from senior stakeholders
- 1 expansion converted or co-pitched
Verification path
1Self-assess + multi-reviewer calibration
2Submit portfolio
3Manager + ODC Lead + Head of D3 sign-off
PM4ODC Lead
When opportunity arises
What to demonstrate
- Demonstrated client trust at C-level
- Track record of account growth during PM4 tenure
- Successful hand-off of previous team to a successor
Portfolio (CORE)
- Client C-level reference
- Account growth data (revenue, headcount, scope)
- Successor in role for 3+ months with stable metrics
Verification path
1ODC Lead opening or new account matched
2Pass ODC Lead assessment (separate framework)
3Head of D3 + Delivery Director sign-off
Alternative tracks — when the PM path isn't the best fit
Delivery Manager
Internal-facing. Run the inside of the engagement so a client-facing PM runs the outside.
When to consider
- You freeze in front of clients but run a tight team
- English is a hard ceiling, execution is strong
- Your account has a separate Account Manager
Profile fit
Strong: Delivery, Risk, Team Leadership, People Mgmt (L4+)
Relaxed: Client Comm, Account Health (L2–L3 OK)
Relaxed: Client Comm, Account Health (L2–L3 OK)
Comp parity
Same band
Reversible
Yes, 1×/18mo
Solution Architect
Technical-facing. For PMs with engineering background who prefer architecture depth over client work.
When to consider
- You're drawn into design discussions and add value
- Background is engineering, energized by tech
- Client Comm gap is preference, not skill
Profile fit
Strong: Tech Literacy (L4+), Risk technical (L4+)
Relaxed: Client Comm, Team Leadership, People Mgmt
Relaxed: Client Comm, Team Leadership, People Mgmt
Comp parity
Engineering ladder
Reversible
Yes, level resets
How to move between tracks
1. Declare interest
2. 1-on-1 with supervisor
3. 1-sprint shadow + 1-sprint contribute
4. Decide within 30 days
FAQ for PM role
FAQ items will be added as questions surface from the team.
Self-assessment
Self-assess your PM level
Score yourself on each dimension and get a live radar chart of your competency profile. Scores stay in your browser — never saved or shared.