Benchmark report

Website Performance Benchmark

What South African institutional websites actually cost their visitors, measured rather than asserted.

In field — no findings claimed

Status: In field — no findings claimed
Expected: 28 February 2027
Reading time: 16 minutes
Subjects: Digital transformation, Higher education, Government

Method and sample

Method

Automated measurement of Core Web Vitals, accessibility violations and page weight across a fixed page set per institution, run from a South African network location on a throttled mobile connection, repeated on three separate dates. Tooling, throttling profile and page selection rule published before measurement.

Sample

Target: 26 public universities, 20 national and provincial departments, and the 8 metropolitan municipalities. Fixed page set per institution: homepage, primary service or admissions page, contact page, and one deep content page selected by a stated rule rather than by the assessor.

Key points

Measured from a South African network location, because measuring from Frankfurt flatters everyone.
Throttled mobile connection, which is how most visitors arrive.
Three separate measurement dates, so a good day is not reported as a standard.
Page selection by stated rule, not by assessor preference.

Programme detail

Status: in field

Measurement opens November 2026. No figures are published in advance, and no institution will be named in a way that presents a single bad measurement as a settled fact.

Why local measurement matters

Performance measured from a well-connected European data centre bears little relationship to what a visitor in Mangaung experiences on a throttled mobile connection. Any benchmark that does not state its measurement location is not measuring the thing that matters.

What will be reported

Distribution of Core Web Vitals by sector; accessibility violation counts by category; median page weight; and the relationship, if any, between platform choice and outcome. Where the relationship is weak, that will be reported as weak.

Citations and downloads

Sources and references

[1]Core Web Vitals, GoogleThe performance metric definitions applied.
[2]Web Content Accessibility Guidelines 2.1, Level AAThe accessibility standard applied.

Download

Measurement protocol (PDF)

Tooling, throttling profile and page selection rule, published before measurement.

Request the document

Sent by email within one business day. No registration, no marketing sequence.

Related frameworks

On what this page does not claim

This report is in field. No findings are available and none are claimed here. The method and sample above are published in advance so that the measure cannot be adjusted after the results are known.

The Diagnostic Hierarchy

Why most communication briefs describe a symptom, and the nine-level order in which to look for the cause.

11 min read

Guide

AI Readiness Before Adoption

A governance-first guide for executives who must hold an AI position before they hold an AI budget.

14 min read

Benchmark report

South African AI Readiness Report

A flagship annual benchmark of AI readiness across universities, government departments, municipalities and mid-market firms.

22 min read

Book a Discovery Session

The most useful first step is a discovery conversation.

A structured session that produces a written problem statement, whether or not it leads to an engagement. If the problem is not one we should be paid to solve, we will say so.

Book a session Send an enquiry instead

Enquiries acknowledged within one business day, always.