Distinguish missing values from genuine zeros
Handle blanks and zeros according to their business meaning, preserve uncertainty and avoid silently changing totals or averages.
4 min readRead guideTHE ACCURALYTIX LEARNING LIBRARY
Better questions. Clearer decisions.
A practical learning library for trustworthy data, understandable reporting and thoughtful business decisions.
72 guides
Saved on this browser. No account required.Measure completeness against the population you expected
Check completeness at the record and field level, explain the expected population and investigate missing-data patterns before reporting a percentage.
4 min readRead guideResolve duplicate entities without deleting useful history
Distinguish duplicate records from legitimate repeated events and make entity matching reviewable before combining or removing information.
4 min readRead guideWrite validation rules that respect legitimate exceptions
Use format, range and relationship checks to identify records for review without confusing valid-looking values with verified truth.
4 min readRead guideShow data freshness and reporting periods separately
Help report readers distinguish the period measured, the source update time and the dashboard refresh time before acting on an apparent change.
4 min readRead guideChoose identifiers that survive a system handover
Keep customer and transaction relationships intact by specifying identifier scope, preserving old keys and testing migration mappings.
4 min readRead guideKeep units and currencies visible during an import
Import measurements and money with explicit units, conversion rules and original values so apparently plausible totals remain interpretable.
4 min readRead guideSet reporting cut-offs before comparing time periods
Define event clocks, time zones and late-arrival rules before assigning records to days, weeks or reporting periods.
4 min readRead guideBuild a category taxonomy people can maintain
Design categories with operational definitions, ambiguous-case examples and a controlled change process while preserving original source labels.
4 min readRead guideCheck relationships between records
Find broken references between customers, orders and other linked records while distinguishing timing differences from genuine relationship errors.
4 min readRead guideInvestigate outliers before correcting them
Distinguish unusual observations from verified errors by checking comparable populations, original evidence and the effect on reported measures.
4 min readRead guideHandle late-arriving data without silent revisions
Build an explicit revision policy that keeps event periods, arrival dates and previously published totals traceable when late records appear.
4 min readRead guideMake spreadsheet imports repeatable
Control spreadsheet structure, parsing and reconciliation so a new title row or locale change cannot silently shift imported meaning.
4 min readRead guideAgree a source-change contract
Create a practical agreement for field, status and meaning changes so downstream reports can assess compatibility before new data arrives.
4 min readRead guideTriage quality issues by their decision impact
Prioritise data issues by the decisions, populations and values they can distort rather than by error counts alone.
4 min readRead guideUse a data-quality score without hiding exceptions
Design quality summaries with explicit denominators, dimensions and critical exceptions so an attractive average cannot override a material limitation.
4 min readRead guideValidate accuracy with a source-based spot check
Check selected values against authorised supporting records, document selection and discrepancies, and keep conclusions within the sample’s evidence.
4 min readRead guideCompare consistency across sources without forcing agreement
Reconcile source differences through concepts, scope and timing before treating every unequal total as an error to eliminate.
4 min readRead guideSeparate revenue, bookings and cash collections
Define signed commitments, recognised revenue and receipts on separate clocks, then explain their legitimate differences through a traceable bridge.
4 min readRead guideDefine an active customer before counting one
Choose a meaningful qualifying event, identity unit and observation window so active-customer counts describe the behaviour readers expect.
4 min readRead guideBuild a retention metric with a clear cohort
Compare customer return behaviour at the same cohort age with explicit entry events, return windows and eligibility rules.
4 min readRead guideDistinguish gross and net revenue in a report
Explain the deductions and timing behind gross and net sales views while keeping management labels distinct from formal accounting policy.
4 min readRead guideChoose a conversion window that fits the journey
Define conversion opportunity from a shared entry event and show how observation windows change measured outcomes without implying causation.
4 min readRead guideCalculate a weighted average without hiding the weights
Calculate pooled averages from appropriate weights and disclose what those weights represent so small groups do not receive accidental equal influence.
4 min readRead guideChoose between the mean and the median
Choose a summary that fits the decision, examine distribution shape and keep valid long cases visible when describing typical service times.
4 min readRead guideAggregate ratios using the underlying counts
Pool conversion and other rates from compatible numerators and denominators while preserving group counts and composition effects.
4 min readRead guideDefine inventory turnover consistently
Choose matching inventory and throughput bases, a stated averaging method and a clear period before interpreting stock turnover.
4 min readRead guideMeasure capacity utilisation with an honest denominator
Distinguish paid, available and productive hours so capacity utilisation supports planning without hiding absence, rework or delivery constraints.
4 min readRead guideDefine a backlog by status and age
Create a reproducible backlog snapshot using explicit eligibility statuses, age clocks and treatment of blocked or administratively open work.
4 min readRead guideMake a gross-margin definition reproducible
Specify revenue, cost scope and matching periods so gross-margin comparisons can be reconstructed and classification changes remain visible.
4 min readRead guideTreat refunds according to the reporting question
Keep refund events linked to original orders while choosing order-cohort or adjustment-period views that answer distinct reporting questions.
4 min readRead guideCount customers across several systems
Count a defined customer entity across billing, CRM and service tools using reviewed mappings, clear commercial scope and visible unresolved identities.
4 min readRead guideExplain recurring revenue without counting every receipt
Define a recurring commitment metric with explicit inclusion, monthly normalisation and cancellation rules while keeping receipts and recognised revenue separate.
4 min readRead guideCompare funnel stages using a shared entry cohort
Follow the same eligible entrants through defined stages and windows so funnel ratios do not mix unrelated period activity.
4 min readRead guideCompare regions on a sensible basis
Choose regional comparison bases deliberately, separating total scale, exposure, currency and customer mix rather than hiding them in a single ranking.
4 min readRead guideBuild a metric dictionary people can actually use
Write concise metric definitions with reproducible formulas, populations, clocks and worked boundary cases that connect directly to published reports.
4 min readRead guideSeparate correlation from an explanation of cause
Separate observed association from causal explanation, compare credible alternatives and choose evidence that could distinguish them.
4 min readRead guideSegment a result before explaining the average
Explain an aggregate result by examining subgroup rates, subgroup weights and stable definitions before attributing a change to performance.
4 min readRead guideInspect seasonality before interpreting a trend
Check recurring calendar patterns, exposure and comparable trading periods before interpreting adjacent observations as a sustained trend.
4 min readRead guideDescribe what a small sample can and cannot show
Use small samples to investigate experiences and refine questions while keeping population claims, uncertainty and recruitment limits explicit.
4 min readRead guideRecognise survivor bias in a success report
Recognise when a success report excludes people who left, failed or disappeared, and rebuild the relevant starting population.
4 min readRead guideCheck how a sample was selected
Review recruitment, coverage, response and eligibility to understand whose experience a sample can describe.
4 min readRead guideLook for a confounding change
Check simultaneous audience, operating and measurement changes before attributing an observed result to one intervention.
4 min readRead guideUnderstand regression toward the mean in a follow-up
Interpret follow-up results after an extreme observation by checking selection, ordinary variability and a credible comparison.
4 min readRead guideUse sensitivity analysis to expose a fragile conclusion
Test whether a recommendation changes when uncertain inputs move within justified ranges, and identify the assumptions worth resolving.
3 min readRead guideBring the base rate into an alert review
Use event prevalence and a complete alert confusion table to distinguish detection rate from the chance that an alert is genuine.
4 min readRead guideReport absolute and relative changes together
Show counts, rates, percentage points and relative movement with a consistent baseline and denominator.
4 min readRead guideExplain uncertainty instead of adding false precision
Communicate ranges, assumptions and unresolved evidence without presenting unsupported decimal places or invented statistical intervals.
4 min readRead guideBacktest a forecast against information available at the time
Evaluate forecasts using frozen information sets, consistent horizons and transparent error measures instead of later corrected knowledge.
4 min readRead guideKeep missing-data uncertainty visible in an analysis
Keep absent groups and unresolved values visible, test defensible bounds and explain when missing data could reverse the decision.
4 min readRead guideAvoid searching many comparisons for one attractive result
Distinguish exploratory segment searching from a planned test, disclose the search scope and seek independent confirmation before strong claims.
4 min readRead guideBuild a causal hypothesis as a testable chain
Turn a proposed cause into an observable sequence with timing, competing explanations and evidence that could contradict the mechanism.
4 min readRead guideCombine qualitative context with quantitative observations
Use interviews and operational records together while preserving each source’s scope, timing and limits on causal interpretation.
4 min readRead guideSet an action threshold using consequences
Choose alert and action thresholds by comparing consequences, review capacity and uncertainty rather than selecting a convenient score.
4 min readRead guideRead a sales pipeline for health rather than volume
Assess pipeline evidence, stage criteria, ageing and next steps before using opportunity totals to guide a sales review.
4 min readRead guideBuild an invoice-ageing view with clear dates
Build invoice ageing from an explicit reporting date, due-date convention and reconciled open balances, with disputed and unapplied items visible.
4 min readRead guideReconcile promised cash with collected cash
Reconcile commitments, invoice balances and verified receipts with explicit cut-offs and allocations instead of treating promises as cash.
4 min readRead guideCreate an inventory exception report people can act on
Create an inventory exception queue with trustworthy units, dated stock positions, demand assumptions and specific actions for each risk.
4 min readRead guideCompare staffing demand and available capacity
Compare demand with usable skill-specific hours, availability and timing rather than treating headcount as interchangeable capacity.
4 min readRead guideInvestigate delivery lateness by the relevant milestone
Investigate lateness using the promise and completion milestone relevant to the customer, with revised dates and missing events visible.
4 min readRead guideMeasure service quality with a balanced view
Assess service quality with resolution, reopenings, effort and customer context so closure volume does not become the sole target.
4 min readRead guideClassify product returns without oversimplifying reasons
Classify returns with reviewable evidence, multiple contributing factors and an unknown category before choosing a corrective action.
4 min readRead guideRead a support queue using age and complexity
Review support backlog by age, stage and complexity while reconciling arrivals and closures so shrinking volume does not hide neglected work.
4 min readRead guideInvestigate purchase-price variance with context
Investigate purchase-price differences using matched specifications, units, quantities and cost boundaries before judging supplier performance.
4 min readRead guideUnderstand how product mix changes overall margin
Separate within-product margin from sales-mix movement using consistent revenue and cost boundaries and a reconciled aggregate.
4 min readRead guideCompare acquisition spending with a defined cohort
Align acquisition costs with the customers and observation windows they support, and keep attribution uncertainty and delayed outcomes visible.
4 min readRead guideInvestigate a regional performance difference
Investigate regional differences with matched coverage, exposure and customer mix before ranking teams or attributing performance.
4 min readRead guideExplain forecast-versus-actual movement
Explain forecast-to-actual differences with a reconciled bridge that distinguishes assumptions, timing, mix and data revisions.
4 min readRead guideCreate an operational review agenda from the data
Build an operational meeting around material exceptions, decisions and follow-up evidence instead of reading every dashboard measure aloud.
4 min readRead guideDefine unit economics around the actual service unit
Define the actual unit of service, cost boundary and outcome window before interpreting per-unit economics.
4 min readRead guideReconcile finance and operations without forcing one total
Explain legitimate differences between bookings, delivery, billing and cash through an event-based reconciliation rather than forcing equal totals.
4 min readRead guideTrack whether an action changed the intended outcome
Evaluate operational actions with dated hypotheses, implementation evidence, matched outcomes and competing explanations.
4 min readRead guideA SHORTER STARTING POINT
Read the short field notesNeed a quick field note?
Keep the original concise guides and practical resources close at hand.