ICT Contractual Claim · Technical White Paper
ICT Contractual Claim — Technical White Paper
Evidence-led reconstruction of a multi-site telecommunications, network, hardware, SLA and billing dispute, and the governance method that made it reviewable.
Publication boundary
This paper describes a claim that was submitted, not a settlement that was reached. The source record ends with the 13 September 2021 submission at CEO level; it contains no settlement agreement, award or evidence of recovery.
Client and supplier identities, site addresses, individual names and unit rates are excluded. The escalation chronology is described by role rather than by the names recorded in the source. The engagement ran under the law and regulations in force in 2020–2021.
Executive overview
Ten months, twelve disciplines, 774 pages
The engagement concerned a long-running enterprise ICT relationship supporting a distributed care environment — voice services, PBX equipment, wide-area connectivity, MetroLAN, hosted services, switches, wireless access points, cabinets, UPS equipment, cabling and support. By the time KuTh entered the matter the issues were no longer a single billing query.
KuTh was retained on 1 November 2020. The final contractual report, dated 13 September 2021, ran to 123 pages and was supported by 651 pages of annexures. Twelve contributors were recorded across commercial, telecommunications, network, fibre, PBX, Wi-Fi, electrical and systems disciplines. The work culminated in a nine-area contractual claim totalling R23.4 million.
2. Evidence architecture
The annexure record was built from distinct evidence families
Scroll table sideways →
| Evidence family | Scale | Function in the dispute |
|---|---|---|
| Contract / legal record | ~412 pages | Establish accepted scope, prices, service obligations and terms |
| RFO / site specification | ~30 pages | Test what infrastructure and protection items were required and paid for |
| Financial / count evidence | ~18 pages | Anchor pricing and quantity comparisons |
| Performance evidence | ~23 pages | Test network availability and throughput against contracted service |
| Breach support | ~3 pages | Support the historic billing and escalation dispute positions |
| Site-survey evidence | ~99 pages | Corroborate equipment, installation and condition findings physically |
| Corroborating record | ~63 pages | Supporting context and cross-checks |
An enterprise supplier can appear to provide one integrated solution while the legal relationship is actually several quotations, schedules, addenda, acceptance certificates and service-specific obligations. The claim had to identify which document governed each disputed charge.
5. Technical infrastructure findings
The investigation did not assume every outage was a carrier problem
Cable type, pathways, weather exposure, cabinet design, power protection, wireless placement, switch condition and maintenance can all materially affect service stability. Each was examined.
- Cabling and patching. Widespread copper-clad aluminium patch leads, with higher attenuation and DC resistance than full copper and suitability concerns for power-over-Ethernet. Worn or poorly seated crimped connectors gave a plausible physical source of intermittent disconnects. Only seven outdoor permanent links used correct polyethylene-sheathed cable; indoor PVC cable was observed outdoors elsewhere. Weak labelling lengthened fault-finding because technicians could not reliably trace connections.
- Cabinets and environmental exposure. Outdoor cabinets were windowed units without adequate sealing or ventilation, some in direct sunlight with recorded temperatures above 40 °C — treated as relevant to switch and UPS life, not appearance. Conduit entry points were inconsistently sealed, and some cable routes crossed roads using cable not intended for that application.
- Wireless and switching. Indoor access points were recorded outdoors, including units in direct sunlight. Antenna orientation and placement were questioned, with dead zones and missing APs at multiple sites. Fast-Ethernet switching was a constraint where a gigabit design was required.
- UPS and electrical protection. UPS units were found unplugged, missing, damaged or unable to hold charge; at least one site recorded battery-acid leakage and corrosion of the cabinet base plate. The earthing review found a disconnect between RFO requirements, separately billed protection services and the protection actually observed.
11. Claim construction
Four layers, sequenced so none can double-count another
Area 1 — price
The 2015 accepted prices were treated as the applicable base where a later price mechanism was not established. At equal quantities this produced a monthly difference of R109,815.31, applied across 36 months for R3,953,351.16.
Area 2 — quantity
The quantity effect isolated after the price effect has already been taken: R28,313.23 per month over 36 months, or R1,019,276.28. This sequencing is what prevents the same variance being counted twice.
Area 3 — physical presence
From documentary delivery to what was actually found. R50,425.92 per month over nine months, or R453,833.28. This is where the site-survey programme enters the financial model directly.
Area 4 — actual utilisation
Presence was not treated as proof of value. Equipment on site was compared with equipment working and in use: R65,899.52 per month over nine months, or R593,095.68.
11.5 – 11.9
The remaining five mechanisms
The SLA claim dominates at R12,929,514.31 — R538,729.76 per month over 24 months, drawn from the supplier fault-summary workbook and the SLA data available to the review. It was presented separately from the billing and asset-reconciliation claims so its basis was transparent to the executive reading it.
Area 6 addressed earth bars, trip switches and lightning protection, which the report positioned as already within the earlier RFO scope and later charged again under a separate 2018 service: R3,740.03 per month over 49 months. Area 7 concerned equipment specification — outdoor cabinets, trunking and media converters — at R27,174.96 per month over 49 months. Area 8 applied the R17,525.35 MetroLAN upgrade charge over 49 months.
Area 9 applied a 9% average rate to the overpayment position as a separate time-value layer of R2,080,648.65, bringing the consolidated submission to R23,403,296.02.
12. Escalation
A deliberate path, not a single executive event
KuTh began with account-level liaison, targeted visits to the most problematic sites, proposal submissions and a supplier panel presentation, then moved through technical verification, billing-correction submissions and clarification meetings. Between 25 June and 19 July 2021 the matter was escalated above account management to the supplier’s Managing Executive for the relevant market, and then further through senior management.
The later phase included COO-level engagement while the technical, billing and contractual positions were consolidated. On 13 September 2021 the complete package was submitted at CEO level. By that point the dispute had moved out of operational problem-solving and into executive decision-making: one integrated claim rather than a series of disconnected queries that each new account team could restart.
13. Reconciliation controls
Where the source record disagreed with itself
Scroll table sideways →
| Reconciliation point | Effect | Control applied |
|---|---|---|
| Performance history requested vs received | Six months requested; one week of trend data received, plus ~4 weeks of router monitoring | Do not describe the observed week as a six-month measured dataset |
| Supplier site-count verification | An early install-check workbook was superseded as records consolidated | Final claim used the reconciled site-survey and supplier-verification record |
| Media-converter installed count | 177 in the breach summary, 186 in the later claim narrative | Reconcile to the original asset schedule before using it as a definitive fact |
| Corrosion summary arithmetic | The extracted line does not reconcile hours, days and months | Use the ticket-level workbook, not the summary conversion |
| Fault cause-code quality | Some tickets coded as corrosion were closed by reset or assist actions | Analyse root cause and resolution action separately; do not treat the code as fact |
Multiple source systems, evolving site counts and technical datasets were reconciled as separate controls before being consolidated into the executive claim position.
14. Dispute governance method
A reusable structure for complex supplier disputes
Define the dispute perimeter
Service list, contract list, accounts, sites, periods and disputed mechanisms — preventing scope creep and missing service families.
Build an evidence index
One indexed repository linking each proposition to its source, so escalation is reproducible and nothing is searched for twice.
Reconstruct the contract spine
Accepted quotes, addenda, SLAs, RFOs and acceptance certificates — establishing the commercial baseline before any invoice review.
Reconcile the asset chain
Quoted → delivered → on site → working → billed, separating quantity, specification and serviceability problems.
Test performance
Faults, uptime, throughput, monitoring and configuration — determining whether paid-for service became usable service.
Build the billing bridge
Current invoice → corrected invoice → disputed amount → continuing payment, controlling cash flow while preserving the dispute.
Quantify claim areas
Separate principal mechanisms, service-credit calculation and interest, so each can be reviewed independently.
Run an escalation ledger
Owner, request, date, promise, dependency, next escalation — so the supplier process cannot continually restart.
Validate resolution
Credits, invoices, corrected assets, service changes and cash receipt — distinguishing a favourable response from actual commercial closure.
15. Present-day context
Current complaint procedure is governance context, not retrospective law
The underlying engagement ran in 2020–2021 under the legislation then in force. South Africa’s Electronic Communications Act 36 of 2005 remains a central statutory framework, while ICASA’s End-User and Subscriber Service Charter regime has been amended since, including amendments published in 2023 and further amendment regulations in January 2026.
ICASA’s current procedure asks a complainant to lodge first with the service provider and obtain a reference number, allow 14 working days for resolution, then refer an unresolved complaint to ICASA, which treats quality-of-service and billing complaints as categories within that process. Useful as governance context — but it should not be applied retrospectively to change the legal analysis of the 2021 dispute.
17. Assessment questions
If you are carrying a complex ICT dispute
- Can every disputed charge be traced to the document that governs it — the accepted quotation, addendum, SLA or acceptance certificate?
- Does your asset chain run quoted → delivered → on site → working → billed, or does it stop at the supplier’s register?
- Does the fault dataset identify service category, SLA, downtime and resolution action? A service-credit model depends on incident-level evidence.
- Can each claim-area calculation be traced to its source dataset and period basis, so reviewers can test components independently?
- Has live-payment control been separated from the historic claim, protecting continuity without conceding the dispute?
- Does each supplier request have an owner, a due date and a next escalation point? Without that, complex disputes recycle through new account teams.
- Is there a defined escalation path through senior management to executive submission level, for when operational channels cannot resolve the combined position?