Performance Calibration Tool

Graduate Data Engineer · Performance Levels

Fill in the fields below, then save as PDF or print at A2 landscape.
Role

Graduate Data Engineer

Level 1
Unsatisfactory
Low
Individual
Impact
  • Fails to complete assigned SQL or Python tasks even with close guidance and repeated examples.
  • Submits pipeline changes that break existing data quality checks without noticing or flagging the issue.
  • Does not raise blockers when stuck, causing tasks to stall for days without visibility to the team.
  • Makes no effort to understand existing pipeline code before making changes.
Examples
  • Assigned a well-scoped dbt model with a working example; delivered nothing meaningful after two sprints.
  • Pushed a SQL change that introduced NULLs into a downstream fact table without any validation.
Dampeners
  • Was given unclear onboarding with no structured pairing or data platform orientation.
  • Assigned tasks requiring SQL complexity well beyond their current level without scaffolding.
Progression Signal
  • Begins asking for help within a reasonable timeframe when blocked on a query or pipeline task.
  • Completes at least one small, scoped SQL transformation end-to-end without rework.
Business Impact
Impact
  • Data consumers receive unreliable or incorrect data due to unvalidated pipeline changes.
  • Senior engineers spend significant time fixing regressions rather than delivering platform improvements.
Examples
  • Two senior engineers spent a combined day correcting data quality issues introduced in a graduate's PR.
Dampeners
  • Impact is limited given the scope of work assigned; systemic issues may reflect onboarding failure.
Progression Signal
  • Work begins to complete without requiring rework, reducing overhead on the team.
Mid
Individual
Impact
  • Completes some data tasks but with frequent errors requiring senior correction - incorrect joins, missing null handling.
  • Does not follow the team's SQL or Python style conventions despite guidance and examples.
  • Shows limited engagement in ceremonies - not asking questions, not flagging data concerns.
Examples
  • Submitted dbt models consistently without data quality tests despite it being a stated team expectation.
  • Attended standups but never flagged being stuck on a pipeline issue; fell behind without raising it.
Dampeners
  • Recently joined; early patterns may not yet be representative of trajectory.
Progression Signal
  • Begins following the team's SQL conventions with prompting.
  • Starts flagging pipeline blockers in standup rather than waiting to be asked.
Business Impact
Impact
  • Rework and correction overhead falls on senior engineers, slowing data platform delivery.
  • Inconsistent SQL quality contributes to technical debt in the transformation layer.
Examples
  • Three dbt model PRs in one sprint required full rewrites by the reviewer.
Dampeners
  • Business impact remains low given task scope; cost is mainly senior engineer time.
Progression Signal
  • Rework rate drops; PRs begin to land without major SQL or logic corrections.
High
Individual
Impact
  • Consistently resistant to feedback from data reviews; repeats the same SQL mistakes across PRs.
  • Does not engage with learning resources - dbt documentation, orchestration guides, internal runbooks.
  • Creates a negative dynamic in the team through disengagement with data quality discussions.
Examples
  • Received the same code review comment about fan-out join risk five times with no change in behaviour.
  • Declined pairing sessions on orchestration tooling offered by the senior engineer.
Dampeners
  • External factors (personal circumstances, mental health) may be contributing.
Progression Signal
  • Engages constructively with SQL feedback at least once; shows willingness to change approach.
Business Impact
Impact
  • Sustained disengagement begins to affect team morale and data quality culture.
  • Ongoing intervention from senior engineers diverts attention from platform delivery work.
Examples
  • Team lead spent 20% of the week over a month managing this individual's pipeline output.
Dampeners
  • Scope of impact is limited to team level; no direct impact on downstream data consumers.
Progression Signal
  • Senior engineer time required drops as individual begins to self-manage basic pipeline tasks.
Level 2
Development Needed
Low
Individual
Impact
  • Completes small, well-defined SQL tasks but requires step-by-step guidance to do so.
  • Pipeline code is inconsistent - data quality tests are sometimes missing, naming is unclear.
  • Does not flag blockers proactively; waits to be asked about progress in standup.
Examples
  • Completed a simple dbt staging model with close support but could not adapt the pattern independently.
  • Missed adding null checks that were covered in the team's data quality standards guide.
Dampeners
  • Only three months into their first data engineering role.
Progression Signal
  • Begins completing similar SQL transformation tasks with less prompting after seeing the pattern once.
Business Impact
Impact
  • Pipeline work requires review and correction overhead higher than expected for the role.
  • Team must account for graduate dependency in sprint planning; tasks cannot be picked up independently.
Examples
  • Each dbt model averaged two rounds of review comments before merging.
Dampeners
  • Typical for early tenure; mentorship investment now expected to yield returns.
Progression Signal
  • Review round count begins to drop as SQL quality improves.
Mid
Individual
Impact
  • Delivers assigned data tasks but struggles to connect their SQL transformations to broader pipeline context.
  • Questions about data modelling or orchestration are sometimes repetitive, suggesting limited retention.
  • Feedback from data reviews is acted on in the current PR but not carried forward to subsequent work.
Examples
  • Asked the same question about the dbt project structure on three separate occasions.
  • Fixed a review comment about aggregation logic in one model but introduced the same issue in the next.
Dampeners
  • Complex data domain and unfamiliar cloud warehouse may be contributing to retention difficulty.
Progression Signal
  • Starts asking follow-up questions that show understanding of data logic rather than just fixing the immediate comment.
Business Impact
Impact
  • Repeated explanations of pipeline patterns consume senior engineer time that could drive platform improvements.
  • Low carry-forward from feedback slows the pace at which data quality improves.
Examples
  • Senior engineer averaged 45 minutes per day in reactive SQL support rather than structured mentoring.
Dampeners
  • Mentorship investment is expected at this level; the question is whether progress is occurring.
Progression Signal
  • Reactive support time drops as data engineering patterns begin to stick.
High
Individual
Impact
  • Delivers data tasks but shows a pattern of avoiding unfamiliar pipeline territory rather than exploring it.
  • Does not seek stretch SQL or Python opportunities and defaults to the minimum required.
  • Feedback from data reviews is accepted passively but rarely internalised into changed behaviour.
Examples
  • Consistently picks the smallest, most familiar dbt tasks in sprint planning and avoids complex transformations.
  • Acknowledged all code review feedback as noted but SQL patterns unchanged over two sprints.
Dampeners
  • May be lacking confidence rather than motivation; requires a different intervention.
Progression Signal
  • Picks up one unfamiliar pipeline task and completes it with pairing support.
Business Impact
Impact
  • Growth trajectory is below expectation for tenure, creating uncertainty about role fit on the data team.
  • Limited stretch means the graduate is not building breadth to become more self-sufficient on the data platform.
Examples
  • Six months in and still requires the same level of SQL support as at month two.
Dampeners
  • Business impact is low; risk is in longer-term return on mentorship investment.
Progression Signal
  • Takes on and completes a moderately complex data transformation with less support than before.
Level 3
Consistently Delivers
Low
Individual
Impact
  • Completes well-scoped SQL and Python tasks reliably with minimal rework after the first review.
  • Follows the team's dbt conventions, naming standards, and data quality test expectations without reminders.
  • Asks good, focused questions about pipeline logic and data modelling - showing prior investigation.
Examples
  • Delivered three dbt staging model tickets in one sprint, each merging after a single review round.
  • Before asking for help on a join issue, documented what they had tried - saving senior engineer time.
Dampeners
  • Work is still scoped and structured by others; independence on the data platform is emerging but not established.
Progression Signal
  • Begins suggesting SQL or modelling approaches before being told, showing developing initiative.
Business Impact
Impact
  • Contributes reliably to sprint velocity within the scope of their assigned pipeline work.
  • Mentoring overhead is reducing to a sustainable level, freeing senior engineers for higher-value data work.
Examples
  • Delivered 100% of assigned data tickets in three consecutive sprints.
Dampeners
  • Contribution scope is still narrow; business impact is team-local.
Progression Signal
  • Scope of reliable contribution begins to expand incrementally to more complex transformations.
Mid
Individual
Impact
  • Reliably delivers assigned pipeline work and is starting to contribute ideas in data modelling discussions.
  • Actively participates in data reviews - asks clarifying questions about SQL logic and learns from feedback.
  • Shows good learning habits - revisits dbt documentation, follows up on pipeline feedback patterns.
Examples
  • Spotted a potential fan-out in a teammate's aggregation model and raised it constructively in review.
  • Completed a self-directed learning goal on window functions and applied them in their next transformation task.
Dampeners
  • Still dependent on well-scoped tickets; does not yet initiate data model scope definition.
Progression Signal
  • Starts breaking down pipeline tasks independently rather than waiting for them to be fully defined.
Business Impact
Impact
  • Reliable contributor whose pipeline work lands consistently, improving team predictability.
  • Growing engagement in data reviews adds value beyond their own assigned models.
Examples
  • Caught two data logic bugs in review across the sprint that would otherwise have reached downstream consumers.
Dampeners
  • Impact still primarily within the team; limited cross-team or cross-domain contribution.
Progression Signal
  • Begins contributing to data modelling decisions that affect more than their immediate ticket.
High
Individual
Impact
  • Delivers reliably and is beginning to self-direct learning toward identified data engineering gaps.
  • Seeks out stretch pipeline tasks and handles them with appropriate pairing support.
  • Feedback from data reviews is internalised quickly and visibly improves SQL and modelling quality over time.
Examples
  • Identified a gap in their orchestration knowledge, worked through it independently, and applied it in the next sprint.
  • Volunteered to pick up an unfamiliar ingestion pattern and delivered it successfully with one pairing session.
Dampeners
  • Still operating within defined pipeline scope; not yet driving data modelling decisions.
Progression Signal
  • Begins contributing a technical perspective in sprint planning or data model design discussions.
Business Impact
Impact
  • Reliable delivery and growing independence reduce mentoring overhead on senior data engineers.
  • Self-directed growth means the investment in this graduate is compounding across the data platform.
Examples
  • Senior engineer noted they spent half the mentoring time compared to the previous quarter.
Dampeners
  • Business impact still limited to team scope; contribution to cross-team data quality not yet established.
Progression Signal
  • Contributes value that extends beyond their own pipeline tickets to broader team improvement.
Level 4
Leading
Low
Individual
Impact
  • Delivers reliably and begins to support other graduates with SQL and pipeline debugging informally.
  • Contributes meaningfully to data modelling discussions, offering perspectives backed by evidence.
  • Takes ownership of small pipeline workstreams without needing the scope fully defined upfront.
Examples
  • Helped a new graduate work through a complex join issue without being asked.
  • Proposed a cleaner dbt model structure in planning and implemented it successfully.
Dampeners
  • Leadership is informal and emerging; not yet a consistent multiplier on data team output.
Progression Signal
  • Is sought out by peers for SQL or pipeline help, signalling growing data engineering credibility.
Business Impact
Impact
  • Beginning to have a multiplier effect - their work and habits lift those around them on the data platform.
  • Reduced dependency on senior engineers for support tasks creates bandwidth for platform improvement work.
Examples
  • Handled graduate pipeline questions for two weeks, freeing senior engineer time for a critical data model.
Dampeners
  • Scope of multiplier effect is still narrow - team-level only.
Progression Signal
  • Multiplier effect grows as their informal data engineering leadership becomes more consistent.
Mid
Individual
Impact
  • Consistently delivers and actively improves team data practices - documentation, SQL standards, pipeline conventions.
  • Is a go-to resource for newer team members on dbt patterns and data quality approaches.
  • Brings a strong data quality mindset that is visible and positively influences the team.
Examples
  • Rewrote an outdated pipeline onboarding guide that had been causing confusion for new data engineers.
  • Ran an informal knowledge-share on a useful dbt macro pattern they had recently learned.
Dampeners
  • Still at graduate level - breadth and depth in data engineering are developing; not yet a technical authority.
Progression Signal
  • Is described by peers as someone who makes the data team better, not just someone who delivers models.
Business Impact
Impact
  • Contribution goes beyond delivery - actively improving data team effectiveness and shared practices.
  • Their positive data quality habits create a better environment for others to succeed.
Examples
  • Documentation improvements they made reduced data platform onboarding time for the next joiner by an estimated 30%.
Dampeners
  • Impact is still primarily internal to the team.
Progression Signal
  • Impact of their contributions begins to extend beyond the immediate data team.
High
Individual
Impact
  • Operating at the ceiling of what a graduate role can deliver on a data platform - a clear candidate for promotion.
  • Proactively identifies gaps in the team's data quality practices and drives improvement without being asked.
  • Is a visible positive force on data team culture, pipeline quality, and peer SQL learning.
Examples
  • Proposed and led an improvement to the team's dbt model review checklist that reduced review cycle time.
  • Consistently the first to volunteer for stretch pipeline work and consistently delivers it.
Dampeners
  • Still operating within graduate-scoped pipeline context; depth of data architecture influence is bounded.
Progression Signal
  • Conversation should be about promotion timeline, not further development at this level.
Business Impact
Impact
  • Delivering at a level that justifies progression to junior data engineer.
  • Positive influence on data team quality and culture has measurable impact.
Examples
  • Team retrospective feedback consistently cited this individual as a contributor to data team effectiveness.
Dampeners
  • Impact ceiling at graduate level; promotion is the right next step.
Progression Signal
  • Promotion case is clear; delay risks disengagement.
Level 5
Transformative
Low
Individual
Impact
  • Performing well beyond the graduate level in both pipeline delivery and data team contribution.
  • Has independently driven improvements to the team's SQL standards or dbt conventions that are now embedded.
  • Acts as a mentor to other graduates on data quality and pipeline patterns despite being in a graduate role.
Examples
  • Identified a repeated class of null-handling bug in the transformation layer and proposed a pattern to prevent it.
  • Mentored two other graduates through their first complex dbt model tickets.
Dampeners
  • Impact, however exceptional, is still scoped by the nature of graduate pipeline work; the role is the constraint.
Progression Signal
  • Promotion should have already happened or be imminent; this level should not persist for long.
Business Impact
Impact
  • Delivering data platform impact more typical of a junior or intermediate data engineer.
  • Their presence has a measurable positive effect on data team capability and pipeline quality.
Examples
  • Attributed with reducing a class of data quality incidents through a validation pattern they introduced.
Dampeners
  • Exceptional at this level but still bounded; promotion unlocks further data platform impact.
Progression Signal
  • Post-promotion trajectory is expected to continue this rate of data engineering growth.
Mid
Individual
Impact
  • Anomalously strong for a graduate - operating as a capable junior data engineer in all but title.
  • Contributes meaningfully to data model design discussions and drives pipeline improvements cross-team.
  • Sets a data quality standard that raises expectations for the graduate cohort as a whole.
Examples
  • Contributed a well-received data model design proposal to a cross-team schema discussion.
  • Independently delivered a non-trivial ELT pipeline end-to-end with no escalation required.
Dampeners
  • This rating at graduate level is rare and should prompt immediate promotion review.
Progression Signal
  • Promotion is overdue; further delay is a retention risk.
Business Impact
Impact
  • Business value delivered through data pipelines is disproportionate to the role level.
  • Cross-team data contributions create value beyond the immediate team boundary.
Examples
  • A data model improvement they designed reduced duplication of effort for two downstream analytics teams.
Dampeners
  • Exceptional and should not persist at this level; promotion should be the priority response.
Progression Signal
  • Post-promotion, data engineering impact trajectory is expected to accelerate.
High
Individual
Impact
  • Performing at a level that makes their graduate title almost irrelevant - they are a mid-level data engineer by impact.
  • Their SQL quality, pipeline design, and peer mentoring rival those of experienced engineers on the team.
  • Represents an extreme outlier in the graduate data engineering cohort.
Examples
  • Delivered an end-to-end ELT pipeline with no guidance that shipped to production without incident.
  • Their dbt model quality and data review contributions are indistinguishable from senior engineers on the team.
Dampeners
  • Keeping this individual at graduate level is a significant retention risk and a failure of the performance process.
Progression Signal
  • Immediate promotion; the performance system has not kept pace with this individual's data engineering growth.
Business Impact
Impact
  • Delivering outsized data platform value; effectively a force multiplier at a graduate cost.
  • Recognised beyond the team as an exceptional data engineering contributor.
Examples
  • Cited by the engineering manager as one of the top contributors in the quarter across the whole data team.
Dampeners
  • This situation reflects a process failure; action on promotion should be immediate.
Progression Signal
  • Promotion resolves the mismatch; continued data engineering growth expected post-promotion.
raganmcgill.co.uk · Graduate Data Engineer Performance Levels