A diagnostic assessment that analyses a region's performance data and configuration data together to identify what is wrong, where it happens, and why. Cloud-based and requiring no new hardware or software licensing — you simply provide the data.
What we can evidence
Find the cells actually costing you throughput, with the cause attributed, not just the symptom.
Of 7,603 cells assessed, 928 (12.2%) were flagged as downlink-throughput offenders. Root cause was attributed across overshooting (46%), high downlink retransmissions (37%), capacity (12%), MIMO (9%), low CQI (7%) and low carrier aggregation (5%).
Separate genuine capacity problems from configuration problems.
In the same analysis, capacity assessment found roughly 197 of 963 sectors running above 75% busy-hour utilisation — the subset where more spectrum or carriers is the real answer rather than retuning.
The problem it addresses
Operators face increasing demands, hidden problems and stalled progress — performance bottlenecks whose root causes sit in configuration and are expensive to find by manual investigation. Insight is deliberately positioned as a low-risk starting point: it correlates configuration with performance so teams stop guessing.
- A KPI tells you which cell is failing. It doesn't tell you why.
- A configuration audit tells you what drifted. It doesn't tell you who it hurts.
- Insight tells you both — and which one to fix first.
The nine areas analysed
| Area | What the analysis covers |
|---|---|
| Performance & root cause | Throughput, accessibility, retainability, availability, worst offenders, PM × CM correlation |
| Access & mobility | RACH, neighbour integrity, handovers, ANR, TAC/LAC, inter-frequency mobility |
| Coverage & interference | Overshooting, dominance, radio quality, retransmissions, PCI, RSI |
| Capacity & traffic balance | PRB utilisation, congestion, carrier imbalance, load redistribution, growth headroom |
| Spectrum & layer strategy | Dormant spectrum, reselection, steering, carrier aggregation, multi-band utilisation |
| Configuration intelligence | Context-aware baselines, parameter drift, scheduler, MIMO, CQI, timers, feature consistency |
| RF power efficiency | RRU power headroom, PDSCH power optimisation, PA/PB strategy, power control, RF constraints |
| Hardware & asset efficiency | Baseband capability, hardware utilisation, board consolidation, licences, asset reuse |
| 5G & NSA readiness | NSA configuration, SCG mobility, PRACH, MU-MIMO, anchors, feature activation |
Offender cells are classified against explicit, bandwidth-dependent thresholds, for example, a downlink throughput offender at 20 MHz is defined as below 5,000 kbps, and an uplink offender at 20 MHz as below 3,000 kbps.
What you receive
- A detailed technical Insight report covering all nine areas, with worst-case tables down to eNodeB level.
- Root-cause summaries with quantified breakdowns.
- An Insight dashboard giving compliance scores by configuration category, quantified optimisation potential, and geospatial pinpointing of the sites needing attention.
- Prioritised, actionable recommendations ranked by potential impact and implementation difficulty.
- A data-driven optimisation roadmap.
Worked example
Four findings from Insight analyses of live operator networks. Each was invisible to performance data on its own — the cause only became attributable once configuration was correlated against it.
These are diagnostic findings, not post-fix improvement figures. Customers anonymised by region and scale.
Handovers that could never happen
Inter-frequency neighbour gaps across 6,341 eNodeBs. In 111 of them the target carrier is physically on the same site — the UE drops instead of moving one layer down.
Fix type: configuration only.
Accessibility loss with no visible cause
21,263 critical, across 6,481 sites, alongside 3,455 PCI conflicts. PRACH false reception shows up on the KPI as random setup failure — invisible without the configuration side.
Fix type: configuration only.
Capacity already sitting in the racks
Nearly 60% of cells were running on legacy or transitional baseband. Reshuffling boards already installed moves 7,985 of them onto modern hardware — nothing purchased.
Fix type: hardware reuse.
An RF problem that was never RF
One territory carried 53% of all VoLTE abnormal releases. Cause attribution traced the bulk to MME-initiated releases — a core-side fault the RF team had been chasing in the radio domain.
Fix type: core investigation.
Duration
Scoped as a one-off engagement against a specific problematic region. The reference engagement used a two-week analysis window. Delivered by RF and optimisation engineers applying performance/configuration correlation and vendor best-practice knowledge.
Delivers
related
Consulting & Optimisation Services
Diagnose it, fix it with validated scripts, benchmark your autonomy
Service Provisioning for Transport and IP Domains
From resource selection to an approved provisioning output
5G Site Build and Provisioning
One governed workflow from 5G site plan to as-built
4G Site Build and Provisioning
Radio, transport and IP planned together, then reconciled
NetCM
Configuration management you can trust
FlexiPlan
Turn manual planning into governed, repeatable workflows
AiDN Studio
The AI control plane for network operations