Semantic metric layer
One canonical definition per metric, so every answer reconciles.
Decisions from data
AI Analytics & Predictive Intelligence
Natural-language analytics, automated reporting and forecasting on top of governed data.
What this is
Analytics fails on trust, not technology: numbers that disagree between reports, and a queue for every new question. We build a governed metric layer first, then put natural-language querying, automated narrative reporting and forecasting on top of it, so answers are consistent and self-service.
The problem
What this service addresses
If more than one of these is familiar, there is usually a case worth scoping.
- Every new question becomes a ticket for the data team.
- Two reports give two different numbers for the same metric.
- Dashboards show what happened but nobody explains why.
- Forecasting is a spreadsheet maintained by one person.
Scope
What we deliver
Every item below is a deliverable you receive, not an activity we perform.
- Governed metric definitions agreed with the business and implemented once
- Natural-language querying restricted to the governed metric layer
- Automated narrative reporting explaining movements, not just restating figures
- Forecasting with stated confidence intervals and documented assumptions
- Anomaly detection and alerting on the metrics that matter
- Row-level permissions so users see only their own data
Capabilities
What the system does
Natural-language querying
Questions answered against defined metrics and dimensions, with the query shown.
Narrative reporting
Written explanations of what moved, by how much and which segments drove it.
Forecasting
Time-series projections with intervals, backtested against history.
Anomaly detection
Alerts on statistically meaningful deviations rather than arbitrary thresholds.
Permission-aware access
Row and column-level rules enforced in the query layer.
Applied
Where it is used
Each of these is a bounded job with a measurable result, which is how we prefer to scope a first build.
01
Self-service business questions
Managers ask in plain language and get a consistent, sourced answer.
02
Automated management reporting
Weekly and monthly narrative reports generated from governed data.
03
Demand and revenue forecasting
Projections with intervals to support staffing, stock and budget decisions.
04
Operational anomaly alerting
Early warning when a key metric deviates from expectation.
05
Cohort and retention analysis
Standardised cohort views without bespoke analysis each time.
Architecture
How it is put together
An indicative architecture. The final design follows your systems, data boundary and constraints.
-
01
Data sources
Transactional databases, warehouse, product analytics and third-party platforms.
-
02
Modelling layer
Cleaned, conformed tables with tested transformations and documented lineage.
-
03
Semantic layer
Canonical metric and dimension definitions with access rules attached.
-
04
Query interface
Natural language translated into governed queries, with the generated SQL visible.
-
05
Reporting and forecasting
Narrative generation, projections and anomaly detection over modelled data.
-
06
Delivery
Dashboards, scheduled reports and alerts into email, Slack or your existing tools.
Security & governance
Controls we build in
Access, privacy and oversight are part of the build rather than a later hardening phase.
- Querying limited to the governed semantic layer, so answers cannot contradict each other.
- Generated queries shown alongside results, making every number checkable.
- Row and column-level permissions enforced in the query layer, not the interface.
- Forecast assumptions and confidence intervals always stated, never a bare number.
- Metric definition changes versioned, so historical reports remain explicable.
Delivery
Our approach
How we run the engagement
Discovery
Process walkthroughs, data review, constraints and a written scope you sign off before build.
Architecture
Data model, integration boundaries, hosting and the security model, agreed up front.
Proof of concept
A narrow build on your real data against an agreed quality bar, so the go-ahead is evidence-based.
Development
Two-week increments, each ending in a working deployment and a review.
Testing
Automated coverage on critical paths, evaluation sets for AI behaviour, acceptance testing with your users.
Deployment
Staged rollout into your infrastructure with monitoring and a tested rollback path.
Monitoring & support
Quality, cost and latency tracking after launch, with handover documentation or an ongoing retainer.
Proven work
Products we have built and sold
We are product builders as well as a services team. These are our own commercial platforms, not client references.
Digital sales & support agent
SIZH Virtual Agent
A virtual digital agent for sales, support and reception — trained on your business, live on WhatsApp, web, voice and email.
View product Mobility platformRideOn
A ride-hailing and fleet management platform, now with agentic dispatch, driver support and demand intelligence.
View product Multivendor marketplaceUniBooker
A configurable multivendor service-booking marketplace with agentic onboarding, matching and support.
View product
Questions
AI Analytics & Predictive Intelligence — FAQs
The questions we are asked most often about this service.
Can it just query our database directly?
We strongly advise against it. Unconstrained text-to-SQL produces confident, wrong answers and ignores permissions. A governed metric layer is more work up front and the reason the answers can be trusted.
Do we need a data warehouse first?
Not always. For a focused set of metrics we can model directly over your transactional database with care. Many sources or heavy query volume make a warehouse the better foundation.
How accurate is the forecasting?
We backtest against your history and report error ranges rather than claiming a figure. Where your data is too sparse or volatile to forecast usefully, we say so.
Will it replace our BI tool?
No. It complements it. The metric layer can feed your existing dashboards while adding natural-language querying and narrative reporting.
Related
All AI First services
Services that pair with this
Machine Learning Development
Predictive and classification models trained on your data, evaluated honestly and deployed as services.
Read moreNatural Language Processing (NLP)
Classification, extraction and analysis across the text your business already generates.
Read moreArtificial Intelligence Development
Custom AI applications built around your data, your systems and a measurable quality bar.
Read more