Menu
For boardsServicesProblemsFrameworksHow we workInsightsAboutAssess AI readinessBook a consultation
Service · Foundation

Data and business analytics

Lemur Media advises on data and business analytics. Lemur Media brings order to the path from data sources to decisions: metric definitions, metadata and a data model that supports reporting and AI solutions.

Book a consultationProblem library

When does a company need data advisory?

A company needs data advisory when departments calculate the same metrics differently and the data is not fit to feed AI solutions.

  • The company maintains many reports, yet decisions are made without reference to data.
  • Departments use different definitions of the same metrics.
  • The data is not fit to feed AI solutions.

What does data advisory at Lemur Media cover?

Lemur Media prepares a metrics glossary, a Data Value Chain from source to decision and a plan to improve data quality.

  1. 01Data Value ChainSources, quality, metadata, model, analytics and data products.
  2. 02Metrics glossaryOne definition for each metric, with formula, owner and frequency.
  3. 03Cohort analyticsCustomer retention and value over time instead of period averages.
  4. 04Data for AITaxonomies, metadata and permissions for selected use cases.

What outputs does the board receive?

OutputContents
Data mapSources, flows and owners
Metrics glossaryDefinitions, formulas, owners, update frequency
Gap listGaps blocking analytics and AI, with priorities and effort estimates

How does the Diagnostic work?

The Diagnostic takes 2⁠–⁠3 weeks. Scope and price are agreed before the contract is signed and do not change during the work. It ends with a recommendation on next steps, which the board can carry out with us, on its own or with another partner.

  1. STAGE 1Interviews and dataInterviews with the board and 3⁠–⁠5 teams, a document review and data access within the scope agreed in the non-disclosure agreement.
  2. STAGE 2AnalysisAssessment using the Data Value Chain method, testing hypotheses against data, preliminary conclusions discussed with the client's lead contact.
  3. STAGE 3RecommendationsA report, a 90-day roadmap and a board meeting at which we recommend a decision: pilot, programme, interim management or closing the topic.

How much time does this require from the client's team?

  • Person responsible for data or analytics
  • Access to system and data warehouse documentation
  • List of the most important management reports

Which indicators show progress?

  • Share of management metrics with a single definition and owner
  • Time needed to prepare a management report
  • Share of data for the selected AI use case with complete metadata

When is this service not the right choice?

  • Projects that only require an integrator to implement a BI tool or data warehouse.

Related problems from the library

Data and analyticsDelegation phase · Coordination phase "We have dashboards, but decisions are made on gut feeling"Revenue PLN 40m–1bn · 50+ employees
E-commerce and D2C, B2B and professional services, Manufacturing and distribution, Media and publishers · Around break-even, Stable positive EBITDAMethod: Data Value Chain + Operating model Data and analyticsDelegation phase · Coordination phase "Our data is not fit for AI"Revenue above PLN 40m · 50+ employees
Media and publishers, E-commerce and D2C, B2B and professional services, B2B SaaS · Around break-even, Stable positive EBITDAMethod: Data Value Chain + LEMUR AI Readiness Index Data and analyticsDelegation phase · Coordination phase "Departments calculate the same metrics using different definitions"Revenue PLN 40m–1bn · 50⁠–⁠999 employees
B2B SaaS, E-commerce and D2C, Marketplaces and platforms, B2B and professional services · Around break-even, Stable positive EBITDAMethod: Data Value Chain Data and analyticsDelegation phase · Coordination phase "The board report takes a week to compile by hand"Revenue PLN 40m–1bn · 50⁠–⁠999 people
Manufacturing and distribution, B2B and professional services, E-commerce and D2C · Around break-even, Stable positive EBITDAMethod: Data Value Chain Data and analyticsCoordination phase "We don't have a single view of the customer"Revenue above PLN 40m · 250+ employees
E-commerce and D2C, Media and publishers, Marketplaces and platforms · Around break-even, Stable positive EBITDAMethod: Data Value Chain + Revenue Engine Map Data and analyticsCoordination phase · Renewal phase "We built a data warehouse that nobody uses"Revenue above PLN 200m · 250+ employees
Manufacturing and distribution, E-commerce and D2C, B2B and professional services · Stable positive EBITDAMethod: Data Value Chain

Full problem library →

Materials in the Insights section

Guide · 4 min

Data quality: dimensions, measurement rules, thresholds and accountability

Data quality problems usually only surface in board reports or in wrong decisions. The article describes six dimensions of data quality, how to record rules and thresholds with a worked example, the split of accountability between the business and technical teams, and common mistakes.Read →Template · 4 min

A one-page management report: template, indicators and review cycle

A management report exists to support decisions and needs more than a description of results. The article describes the structure of a report that fits on one page, rules for choosing indicators, a template with calculated variances against target, how a monthly review runs, and common mistakes.Read →Template · 4 min

Metrics glossary: definition fields, example and change process

A metrics glossary is an agreed set of indicator definitions in which every metric used by the board has one definition, one formula and one owner. The article contains a list of definition fields, a completed example for the metric "active customer" with a calculation of the effects of different definitions, the process for introducing changes, and common mistakes.Read →Guide · 4 min

Data governance in a mid-sized company: roles, rules and a 90-day plan

Data governance is the way decisions about data are made: who is accountable for it, under what rules, and how they are enforced. The article presents the DAMA-DMBOK definition, a minimum set of roles for a company with 50⁠–⁠249 employees, a calculated workload, a 90-day implementation plan and common mistakes.Read →Analysis · 4 min

Data mesh: four principles and when it makes sense for a Polish company

Data mesh is an approach to the architecture and organisation of analytical data in which responsibility for data moves from a central team to business domains. The article describes the four principles according to Zhamak Dehghani, the conditions under which the approach makes sense, a worked example of a request queue and common implementation mistakes.Read →Guide · 4 min

Customer Data Platform (CDP): definition, uses and business case

A Customer Data Platform is software that combines customer data from many systems into a single persistent profile available to other tools. The article describes the CDP Institute definition, how a CDP differs from a CRM and a data warehouse, a worked business case for an online shop and common implementation mistakes.Read →Guide · 4 min

Semantic layer: one metric definition in every tool

A semantic layer is where a company defines its business metrics once, so that all reporting tools and AI assistants use the same definition. The article explains how a semantic layer works, shows a worked example of a metric discrepancy, and describes metric specification and implementation mistakes.Read →Guide · 4 min

Churn analysis: churn rate, causes and a worked calculation

Churn analysis examines how many customers leave, when and why. The article describes the correct way to calculate the churn rate, how to annualise it, the Kaplan-Meier survival curve, the split between voluntary and involuntary churn, and common mistakes.Read →Guide · 4 min

Business intelligence in a company: what BI is and how to measure its value

Business intelligence (BI) is a set of processes, data and tools that turn a company's operational data into reports, indicators and analyses on which the board bases its decisions. We explain the origin of the term, the differences between BI, analytics and data science, the components of a BI system, how to measure its value and adoption, and common mistakes.Read →

Frequently asked questions

Do you implement data warehouses?
We design the model and the requirements. The technical implementation is carried out by the client's team or a selected vendor.

Book an initial consultation

A 30-minute conversation about your company's situation. After the consultation we send a proposed Diagnostic scope, or let you know that the topic is outside our expertise.

Book a consultationProblem library