Data and analytics

Specialist agents · 21

• Source inventoryanalysis

  Maintains the inventory of data sources (CRM, ERP, spreadsheets, exports, APIs) with origin, owner, refresh frequency and sensitivity, and flags sources with no owner, no documentation or disconnected.

• Data integrationproposal

  Proposes ingestion and sync changes in the data warehouse (loads, schedules, field mappings, retries) so CRM, ERP and other sources arrive complete and on time.

• Data qualityproposal

  Detects missing values, duplicates, invalid formats and statistical anomalies in warehouse tables and proposes bounded validation rules and fixes.

• Metric definitionsproposal

  Maintains the metrics dictionary with formula, scope, period and single source, and proposes canonical definitions in the warehouse when two reports compute the same thing differently.

• Master dataproposal

  Proposes normalization of master entities (customers, products, suppliers, locations): shared keys, duplicate merging and value catalogs across CRM, ERP and the warehouse.

• Business analysisanalysis

  Answers business questions with analysis of verified data, explaining causes of variance and explicitly stating evidence limits and possible biases.

• Forecasts and scenariosanalysis

  Builds forecasts and scenarios (base, upside, downside) from historical series, showing assumptions, uncertainty ranges and sensitivity to key variables.

• Reports and dashboardsanalysis

  Designs and delivers reports and visualization views (dashboards, periodic summaries) only on validated data and defined metrics, tailored to each audience and schedule.

• Lineage and traceabilityanalysis

  Reconstructs the lineage of each figure, from report back to original source and intermediate transformations, to audit results and assess the impact of upstream changes.

• Data lifecycleproposal

  Prepares retention, archiving and purge policies in the warehouse based on legal obligations and usage value, proposing deletions only with prior authorization and logging.

• Data modeling and architectureproposal

  Designs the dimensional models (facts, dimensions, grain) of the data warehouse and proposes the analytics stack architecture, balancing scalability, latency and cost for an SME or startup.

• Data privacy and complianceanalysis

  Classifies personal and sensitive data used in analytics and checks that its uses comply with GDPR, Spanish LOPDGDD, CCPA and sector regulation, flagging legal bases, minimization and retention periods.

• Data access governanceproposal

  Proposes roles, permissions and data access controls across departments following least privilege, and prepares the proposal for owner validation.

• Product analyticsanalysis

  Defines product event tracking plans and analyzes funnels, activation, retention and usage cohorts to explain where users drop off and which behaviors predict retention.

• SaaS metrics and investor reportinganalysis

  Calculates ARR, MRR, churn, NRR, CAC, LTV and payback using standard definitions and reconciles figures for board and investor reporting.

• Experimentation and A/B testinganalysis

  Designs experiments and A/B tests, calculates sample sizes and minimum duration, and evaluates statistical significance and adverse effects of results.

• Marketing attribution and revenue analyticsanalysis

  Models multichannel attribution and analyzes pipeline and stage and channel conversion by combining CRM and marketing data to assign credit to each channel.

• Segmentation and predictive modelinganalysis

  Builds customer segmentations and predictive models for churn, propensity and scoring, documenting their validation (metrics, bias, drift) before recommending their use.

• Financial and operational analyticsanalysis

  Analyzes margins and profitability by customer, product or line and reconciles ERP data with management reports to detect mismatches and margin leakage.

• Analytics intake and self-service enablementanalysis

  Prioritizes departments' data requests by value and effort and prepares guides for self-service use of dashboards to reduce repetitive demand.

• AI data readinessproposal

  Assesses the suitability, representativeness, bias and documentation of data used by AI models and agents, and proposes warehouse adjustments to prepare it.