Corporate & Business · NPO & Social Impact · Procurement & Sourcing
The capital cost fell 14.75%. The lifecycle cost fell 1.13%.
A client should be able to explain who entered the process, why they qualified, how their claims were tested, what information they received, how conflicts were handled, how capability and price were compared, and why the final decision can be defended. Confidence in a tender is engineered into the process — known suppliers, referrals, presentations and headline prices do not replace verified evidence.
- 14.75%
- reduction in capital cost
- 1.13%
- reduction over 36 months
- 0
- tender points for a referral
- 15
- locations validated on site
Proof context: An anonymised multi-site infrastructure tender
The principle
A material tender is a governance process, not a quotation exercise
The client should be able to explain who entered the process, why they qualified, how their claims were tested, what information they received, how conflicts were handled, how capability and price were compared, and why the final decision can be defended.
A supplier is not treated as suitable because it is the incumbent, was nominated by the client, is known to KuTh, or was referred by another party. Every supplier must survive the evidence appropriate to its role in the process. Confidence in a tender is engineered into it — known suppliers, referrals, presentations and headline prices do not replace verified evidence, equal treatment and documented judgement.
The commercial result
Why the headline saving was not the real one
Scroll table sideways →
| Commercial measure | Prior position | Revised position | Movement |
|---|---|---|---|
| Capital cost — ex VAT | R16,708,311.94 | R14,244,277.27 | R2,464,034.67 lower — 14.75% |
| Monthly recurring / support | R223,310.37 | R283,964.47 | R60,654.10 higher — 27.16% |
| 36-month nominal total | R24,747,485.26 | R24,466,998.19 | R280,487.07 lower — 1.13% |
The 36-month figure is a simple ex-VAT illustration — capital cost plus 36 months of recurring cost. It excludes financing, escalation, time value of money, tax effects, unpriced risk, change orders, credits and early-termination effects, and it is not a realised-savings claim. A 14.75% reduction in upfront capital did not produce a 14.75% lifecycle saving: higher recurring cost reduced the simple 36-month advantage to roughly 1.13%.
Process safeguards
The controls that make the figure defensible
Scroll table sideways →
| Safeguard | What it means |
|---|---|
| Controlled market | The incumbent is Control 1 and one client-nominated supplier is Control 2. Both establish benchmarks; neither receives automatic preference |
| Zero-point referrals | Market, KuTh-sourced and referred suppliers may challenge the controls. Referral origin is recorded for transparency, but referral itself earns zero qualification or tender points |
| Qualification before reliance | Legal and compliance standing, experience, technical people, delivery capacity, financial ability, quality and regulatory controls and ethics are tested before a supplier is relied upon |
| References and premises verification | References are independently checked. For material tenders, KuTh may visit bidder premises to confirm that represented people, systems, service infrastructure, stock and operating capability exist in practice |
| Testing and client-site surveys | Samples, demonstrations, equipment testing and proof-of-concept are used where documents are insufficient; site surveys test whether a proposal fits the actual operating environment |
| Ethical independence and equal information | Conflicts, referral interests, gifts and inducements are declared; material bidder information and clarifications are controlled; recusals are documented |
| Scoring and QA | Criteria and weights are fixed in advance where practicable, score rationales are recorded, calculations are locked and arithmetic and denominator logic is independently checked |
| Lifecycle value and contract conversion | Scope, capex, recurring cost, term, escalation, implementation and exclusions are normalised. Awarded promises are carried into the contract and checked after implementation |
What the source record taught
Two errors found in the historical workbook, published rather than quietly corrected
One intermediate difference formula omitted an installation and cabling cost column, producing a larger apparent reduction than the difference between the reconciled total-capex values. That is precisely the kind of error that can change a recommendation if a procurement model is not independently checked — and it is why summary movements must equal the difference between controlled underlying totals, not a separate formula.
The historical technical scorecard also contained arithmetic and denominator inconsistencies: one cabinet score line showed a total denominator that did not align with the number of scored criteria. The lesson is not that scoring is unreliable. It is that scoring needs control — locked calculations, recorded rationales, moderation of outliers, and a prohibition on manually entered totals.
Evidence from the engagement
What the process actually covered
Scroll table sideways →
| Measure | Position | Why it matters |
|---|---|---|
| Operating footprint | 15 locations in the final commercial comparison | Multi-site requirements validated at site level, not priced from one generic assumption |
| Initial supplier interaction | 4 supplier names in the chronology | Market engagement before detailed evaluation |
| Detailed technical review | 3 suppliers in the comparative scorecard | The field narrowed. The source does not state every exclusion reason, so none is invented here |
| Site-survey period | 25 January – 2 March 2021 | Suppliers had to test their proposals against the physical environment before finalising |
| Technical modules | Access points, cabinets, firewalls, switches, connectivity, voice and PBX, UPS | Technical evidence extended well beyond price |
| Commercial structure | Capex, recurring support, set-up and connectivity; 36- and 60-month options | Term and cost type had to be normalised before value could be compared |
Evidence and publication boundary
Commercial values are recalculated from the source price-comparison workbook using reconciled totals, not the intermediate difference formula that omitted a cost column. Quotation, negotiated price, contracted price and realised invoice result are not treated as the same thing, and no realised-savings claim is made here.
The source does not state every reason a supplier left the process, so no exclusion rationale is inferred. Client, supplier and site identities are withheld. The 100-point supplier-inclusion model is a KuTh governance control adjusted to category risk; mandatory legal, regulatory or integrity failures override the numerical score entirely.
Supporting documents