Analytics
Analytics explained
Start with a question: Where does an ordering journey fail, which helps people choose, or which enquiries are actually suitable? Then decide which data are needed. A new tool cannot resolve an unclear task. Existing data and direct feedback may already provide useful signals.
Distinguish the type of statement. Descriptive analysis shows what was recorded. Further investigation examines possible explanations. Forecasts estimate future developments under particular assumptions. Recommendations connect those findings to objectives and options for action. These approaches are not a compulsory ladder on which every business must advance to AI models.
Measurement remains a partial view. Technical failures, unobserved journeys, different identifiers and changing definitions can influence the picture. Relationships in data generate hypotheses but do not establish causality on their own. Additional user research or a suitable controlled comparison may be needed to test an explanation.
Creative Engineering connects analysis to a concrete improvement in an idea, content or application. AI may help sort recurring patterns for review. What matters is whether findings are traceable and suggestions useful. Permissions, data quality and expert review belong in the process. We take responsibility for the concept and quality.
Examples
Hypothetical application
A course website records many drop-offs before registration. The team first checks that the journey is measured correctly and examines device types separately. Additional usability testing investigates date selection. A revised selection process is then tested, rather than attributing a particular motive to users based on the drop-off curve alone.
Key Points
- Clarify the question and decision before tools and data collection.
- Distinguish observation, possible explanation and prediction.
- Connect analysis to a testable change.
Practical application
Frame an answerable question and define suitable data and success criteria. Check data quality, analyse relevant groups and state uncertainty. Develop a specific test and assess its outcome with an appropriate method.
Useful measures
Question answered
The analysis transparently explains which statement the data support.
Verifiable data foundation
Definitions, origins, outages and important assumptions are documented.
Tested next step
A resulting change is actually tested and evaluated against appropriate criteria.
Common mistakes
- Equating more stored data with more insight.
- Treating forecasts or attributed results as certain causes.
- Promising a universal tool stack or fixed efficiency gains.
Sources and context
- GOV.UK Service Manual: Using performance data to improve your service
Methodological reference: connect data to a service objective and investigate causes through additional research.
- NIST: What is Exploratory Data Analysis?
Explanation of exploratory data analysis and examining structures, anomalies and assumptions.
Frequently Asked Questions about Analytics
Reporting presents selected results. Analytics also includes investigating a question, checking data and interpretation. A report can be part of that work but does not replace it.
Not necessarily. Scale, sources and tasks determine the technology required. A verifiable simple analysis may be more suitable than a complex model without an adequate data foundation.
No. A different collection method does not automatically resolve missing signals, identity problems or unclear definitions. The conditions for processing data also require separate consideration.
Loading related terms…
All TermsArticles about Analytics
Predictive Analytics in Marketing: Data-Driven Predictions for Better Decisions
How Machine Learning and Predictive Analytics are revolutionizing marketing decisions, from churn prediction and CLV models to demand forecasting.
Behavioral Data Fusion: How Merging Behavioral Data Is Revolutionizing Marketing
Behavioral Data Fusion connects data streams from web analytics, CRM, social media, and IoT into a holistic customer view. Learn why isolated data silos are marketing's biggest blind spot.
Dark Social: How to Measure and Leverage the Invisible 80% of Your Content Reach
Most of your content is shared through private channels, but invisible to analytics. Here's how to use Dark Social strategically.