NPO & Social Impact · Telecommunications & Technology
One package across an estate that varied this much loses money twice
Stranded allowance on the quiet lines, out-of-bundle charges on the busy ones, both caused by the same uniform 2GB default. The remedy was not a downgrade exercise: the right-sized range widened in both directions.
- 30.7%
- equivalent-package migration, modelled
- 40.8%
- with usage right-sizing, modelled
- 88.1%
- of baseline on a single 2GB package
- 84
- connections in the quantified baseline
Proof context: A geographically dispersed mobile-data estate supplied through a reseller
One capability, applied two ways
This engagement and the Corporate Mobile mandate are two examples of the same KuTh mobile capability applied to very different estates — not two separate services.
A corporate multi-entity estate with pooling potential needs a different answer to an NPO data estate bought through a reseller. Both are documented so a reader can go to whichever resembles their own position.
The situation
Eighty-eight per cent of connections on the same package
The organisation ran a geographically dispersed mobile-data estate through a reseller. Of 84 connections in the quantified baseline, 74 sat on the same 2GB package — despite line-level usage running from almost nothing to materially above the allowance.
That concentration is the finding. One package applied across an estate whose actual consumption varies that widely guarantees two simultaneous losses: unused allowance on the quiet lines, and out-of-bundle charges on the busy ones.
The baseline
Where the 84 connections actually sat
Scroll table sideways →
| Package | Connections | Share |
|---|---|---|
| 2GB | 74 | 88.1% |
| 4GB | 4 | 4.8% |
| 5GB | 3 | 3.6% |
| 3GB | 1 | 1.2% |
| 16GB | 1 | 1.2% |
| 24GB | 1 | 1.2% |
Results
Two levers, deliberately separated
Scroll table sideways →
| Result | Classification | What changes |
|---|---|---|
| 30.7% reduction | Modelled / commercially priced | Direct-network equivalent packages against the quantified reseller baseline — the same products, without the reseller margin. |
| 40.8% reduction | Modelled / commercially priced | A usage-led package mix, using lower and higher bundles according to observed consumption. |
| 10.1 percentage points | Analytical uplift | The additional reduction attributable to right-sizing, beyond the pricing change. |
| Signed migration documents | Contract / implementation evidence | Supports that the solution moved past a theoretical quote into an implementation process. |
| Transition disruption | Implementation evidence | Prevents the modelled saving being described as fully realised without clean post-migration billing. |
Why two percentages rather than one
Removing the reseller margin and right-sizing the packages are different kinds of work, and an organisation can do either without the other.
Reporting only the 40.8% would suggest a single negotiation produced it. Reporting only the 30.7% would hide that most of the estate was on the wrong package regardless of who supplied it.
The engagement publishes its own disruption
The retained pack includes signed migration documentation and later line-level upgrade and downgrade schedules — and also records transition disruption during the move.
That is classified as implementation evidence which prevents the modelled saving from being called fully realised. The result therefore separates commercial design from implementation performance, rather than letting a good model stand in for a clean migration.
Which denominator applies
84 connections is the quantified baseline and the denominator for both the 30.7% and 40.8% models.
A later implementation snapshot shows 92 connections. That changed population must not be retrofitted into the earlier calculation — the estate grew, which is a different fact from the saving.
Commercial significance
Different from a corporate pooling exercise
The value came from reconstructing actual data use, removing reseller margin, placing users on smaller or larger packages according to behaviour, and giving the organisation a practical mechanism to manage usage afterwards — central balance and usage visibility, top-up control, package changes, SIM replacement, invoice access, alerts and capped usage through a self-service portal.
The implementation record also shows why migration sequencing, SIM distribution and cut-over governance belong inside the saving method rather than being treated as administrative afterthoughts.
Supporting documents